| {"task_id": "TB-ML-001", "task_type": "objection_handling", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Understood \u2014 bad outsourcing experiences are common. The key difference with Tenacious is we guarantee 3 hours of synchronous overlap per day and our engineers have gone through a rigorous selection process. Not trying to replace anyone, just fill a specific gap.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.724241+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-002", "task_type": "signal_grounding", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "We work with enterprise software teams at your size. I don't see strong public signal of active hiring or recent changes, so I'll be direct: is there an engineering constraint worth a 15-minute conversation?\n\nIf not, no follow-up.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.724384+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-003", "task_type": "objection_handling", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "That's fair. We're not the cheapest option. What Tenacious offers is reliability and overlap \u2014 our engineers are available during your core hours, not just a few. Would love to show you the difference on a small engagement.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.724460+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-004", "task_type": "email_composition", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Context: Tenacious for DataScale Inc\n\nHi,\n\nNoticed 4 engineering roles open since January \u2014 is hiring velocity matching your runway after the Series A?\n\nWe help teams at your stage scale engineering without the 3-month recruiting cycle.\n\nOpen to a 15-minute conversation?\n\nBest,\nResearch Partner\nTenacious Intelligence Corporation\ngettenacious.com", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.724551+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-005", "task_type": "signal_grounding", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "We work with enterprise software teams at your size. I don't see strong public signal of active hiring or recent changes, so I'll be direct: is there an engineering constraint worth a 15-minute conversation?\n\nIf not, no follow-up.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.724693+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-006", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Engineering capacity at DataScale Inc \u2014 quick observation\n\nHi,\n\nYou're clearly scaling aggressively with your AI strategy. We can provide world-class engineers to help. Let's synergize!\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.724822+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-007", "task_type": "objection_handling", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Understood \u2014 bad outsourcing experiences are common. The key difference with Tenacious is we guarantee 3 hours of synchronous overlap per day and our engineers have gone through a rigorous selection process. Not trying to replace anyone, just fill a specific gap.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725043+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-008", "task_type": "signal_grounding", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "We work with enterprise software teams at your size. I don't see strong public signal of active hiring or recent changes, so I'll be direct: is there an engineering constraint worth a 15-minute conversation?\n\nIf not, no follow-up.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725107+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-009", "task_type": "email_composition", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Engineering capacity at DataScale Inc \u2014 quick observation\n\nHi,\n\nYou're clearly scaling aggressively with your AI strategy. We can provide world-class engineers to help. Let's synergize!\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725176+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-010", "task_type": "email_composition", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Context: Tenacious for DataScale Inc\n\nHi,\n\nNoticed 4 engineering roles open since January \u2014 is hiring velocity matching your runway after the Series A?\n\nWe help teams at your stage scale engineering without the 3-month recruiting cycle.\n\nOpen to a 15-minute conversation?\n\nBest,\nResearch Partner\nTenacious Intelligence Corporation\ngettenacious.com", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725237+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-011", "task_type": "competitor_gap_quality", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Looking at public signal from similar-sized companies in your space \u2014 a few have stood up dedicated MLOps functions recently. Wondering how you're thinking about that vs your current setup.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725283+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-012", "task_type": "tone_adherence", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "I understand you had a bad experience. Our offshore team is different \u2014 we provide top talent with cutting-edge skills at competitive rates!", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725328+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-013", "task_type": "email_composition", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Context: Tenacious for DataScale Inc\n\nHi,\n\nNoticed 4 engineering roles open since January \u2014 is hiring velocity matching your runway after the Series A?\n\nWe help teams at your stage scale engineering without the 3-month recruiting cycle.\n\nOpen to a 15-minute conversation?\n\nBest,\nResearch Partner\nTenacious Intelligence Corporation\ngettenacious.com", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725627+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-014", "task_type": "competitor_gap_quality", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Looking at public signal from similar-sized companies in your space \u2014 a few have stood up dedicated MLOps functions recently. Wondering how you're thinking about that vs your current setup.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725668+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-015", "task_type": "signal_grounding", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "You're clearly scaling aggressively with your aggressive AI strategy. Your world-class team needs top talent to succeed. Let's synergize on this!", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725733+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-016", "task_type": "competitor_gap_quality", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Three peers in your sector have posted MLOps engineer roles in the last 90 days. Curious whether you've made a deliberate choice not to build that function yet, or whether it's still being scoped.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725807+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-017", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Request: 15 minutes re DataScale Inc engineering roadmap\n\nHi,\n\nTwo peers in your sector have recently scaled ML teams ahead of product launches. Curious whether that's on your roadmap or already planned internally.\n\nWe work with Series A teams typically 3-4 months post-raise. Happy to share what we've seen.\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725877+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-018", "task_type": "competitor_gap_quality", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Your competitors are clearly ahead of you in MLOps. You're missing critical capabilities that will hurt you long-term.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.725952+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-019", "task_type": "tone_adherence", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "That's a common concern and a fair one. The main difference with Tenacious is timezone overlap and accountability \u2014 we're not just dropping engineers in and walking away. Happy to share how we've handled this for other teams if useful.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726189+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-020", "task_type": "objection_handling", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Appreciate you sharing that. We hear this often. Our value isn't cost \u2014 it's reliability and timezone overlap. Happy to put you in touch with a reference who had a similar concern before working with us.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726258+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-021", "task_type": "competitor_gap_quality", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Your competitors are clearly ahead of you in MLOps. You're missing critical capabilities that will hurt you long-term.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726324+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-022", "task_type": "competitor_gap_quality", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Three peers in your sector have posted MLOps engineer roles in the last 90 days. Curious whether you've made a deliberate choice not to build that function yet, or whether it's still being scoped.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726377+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-023", "task_type": "tone_adherence", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "Appreciate you sharing that context. What typically goes wrong is communication gaps and accountability. We structure engagements to prevent both \u2014 would it help to walk through how?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726419+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-024", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Engineering capacity at DataScale Inc \u2014 quick observation\n\nHi,\n\nYou're clearly scaling aggressively with your AI strategy. We can provide world-class engineers to help. Let's synergize!\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726463+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-025", "task_type": "signal_grounding", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "Noticed 2 open engineering roles \u2014 not a huge signal, but enough to ask whether there's a specific capability you're trying to bring in-house. Happy to share what similar teams have done.\n\nOpen to a quick call?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726619+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-026", "task_type": "objection_handling", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Appreciate you sharing that. We hear this often. Our value isn't cost \u2014 it's reliability and timezone overlap. Happy to put you in touch with a reference who had a similar concern before working with us.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726669+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-027", "task_type": "objection_handling", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "That's fair. We're not the cheapest option. What Tenacious offers is reliability and overlap \u2014 our engineers are available during your core hours, not just a few. Would love to show you the difference on a small engagement.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726731+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-028", "task_type": "signal_grounding", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "Noticed 2 open engineering roles \u2014 not a huge signal, but enough to ask whether there's a specific capability you're trying to bring in-house. Happy to share what similar teams have done.\n\nOpen to a quick call?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.726855+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-029", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Request: 15 minutes re DataScale Inc engineering roadmap\n\nHi,\n\nTwo peers in your sector have recently scaled ML teams ahead of product launches. Curious whether that's on your roadmap or already planned internally.\n\nWe work with Series A teams typically 3-4 months post-raise. Happy to share what we've seen.\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727004+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-030", "task_type": "tone_adherence", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "I understand you had a bad experience. Our offshore team is different \u2014 we provide top talent with cutting-edge skills at competitive rates!", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727092+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-031", "task_type": "tone_adherence", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "That's a common concern and a fair one. The main difference with Tenacious is timezone overlap and accountability \u2014 we're not just dropping engineers in and walking away. Happy to share how we've handled this for other teams if useful.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727325+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-032", "task_type": "email_composition", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Request: 15 minutes re DataScale Inc engineering roadmap\n\nHi,\n\nTwo peers in your sector have recently scaled ML teams ahead of product launches. Curious whether that's on your roadmap or already planned internally.\n\nWe work with Series A teams typically 3-4 months post-raise. Happy to share what we've seen.\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727371+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-033", "task_type": "objection_handling", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "That's fair. We're not the cheapest option. What Tenacious offers is reliability and overlap \u2014 our engineers are available during your core hours, not just a few. Would love to show you the difference on a small engagement.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727411+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-034", "task_type": "tone_adherence", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "That's a common concern and a fair one. The main difference with Tenacious is timezone overlap and accountability \u2014 we're not just dropping engineers in and walking away. Happy to share how we've handled this for other teams if useful.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727450+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-035", "task_type": "signal_grounding", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "We work with enterprise software teams at your size. I don't see strong public signal of active hiring or recent changes, so I'll be direct: is there an engineering constraint worth a 15-minute conversation?\n\nIf not, no follow-up.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727485+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-036", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Engineering capacity at DataScale Inc \u2014 quick observation\n\nHi,\n\nYou're clearly scaling aggressively with your AI strategy. We can provide world-class engineers to help. Let's synergize!\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727519+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-037", "task_type": "signal_grounding", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "Noticed 2 open engineering roles \u2014 not a huge signal, but enough to ask whether there's a specific capability you're trying to bring in-house. Happy to share what similar teams have done.\n\nOpen to a quick call?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727682+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-038", "task_type": "competitor_gap_quality", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Looking at public signal from similar-sized companies in your space \u2014 a few have stood up dedicated MLOps functions recently. Wondering how you're thinking about that vs your current setup.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727802+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-039", "task_type": "signal_grounding", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "You're clearly scaling aggressively with your aggressive AI strategy. Your world-class team needs top talent to succeed. Let's synergize on this!", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.727927+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-040", "task_type": "signal_grounding", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "Noticed 2 open engineering roles \u2014 not a huge signal, but enough to ask whether there's a specific capability you're trying to bring in-house. Happy to share what similar teams have done.\n\nOpen to a quick call?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728056+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-041", "task_type": "objection_handling", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Appreciate you sharing that. We hear this often. Our value isn't cost \u2014 it's reliability and timezone overlap. Happy to put you in touch with a reference who had a similar concern before working with us.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728187+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-042", "task_type": "competitor_gap_quality", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Your competitors are clearly ahead of you in MLOps. You're missing critical capabilities that will hurt you long-term.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728257+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-043", "task_type": "signal_grounding", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "Noticed 2 open engineering roles \u2014 not a huge signal, but enough to ask whether there's a specific capability you're trying to bring in-house. Happy to share what similar teams have done.\n\nOpen to a quick call?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728392+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-044", "task_type": "email_composition", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Request: 15 minutes re DataScale Inc engineering roadmap\n\nHi,\n\nTwo peers in your sector have recently scaled ML teams ahead of product launches. Curious whether that's on your roadmap or already planned internally.\n\nWe work with Series A teams typically 3-4 months post-raise. Happy to share what we've seen.\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728445+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-045", "task_type": "competitor_gap_quality", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Your competitors are clearly ahead of you in MLOps. You're missing critical capabilities that will hurt you long-term.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728482+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-046", "task_type": "tone_adherence", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "That's a common concern and a fair one. The main difference with Tenacious is timezone overlap and accountability \u2014 we're not just dropping engineers in and walking away. Happy to share how we've handled this for other teams if useful.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728515+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-047", "task_type": "objection_handling", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Appreciate you sharing that. We hear this often. Our value isn't cost \u2014 it's reliability and timezone overlap. Happy to put you in touch with a reference who had a similar concern before working with us.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728546+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-048", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Engineering capacity at DataScale Inc \u2014 quick observation\n\nHi,\n\nYou're clearly scaling aggressively with your AI strategy. We can provide world-class engineers to help. Let's synergize!\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728592+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-049", "task_type": "competitor_gap_quality", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Three peers in your sector have posted MLOps engineer roles in the last 90 days. Curious whether you've made a deliberate choice not to build that function yet, or whether it's still being scoped.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728768+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-050", "task_type": "tone_adherence", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "Appreciate you sharing that context. What typically goes wrong is communication gaps and accountability. We structure engagements to prevent both \u2014 would it help to walk through how?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728820+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-051", "task_type": "tone_adherence", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "I understand you had a bad experience. Our offshore team is different \u2014 we provide top talent with cutting-edge skills at competitive rates!", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728869+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-052", "task_type": "email_composition", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Context: Tenacious for DataScale Inc\n\nHi,\n\nNoticed 4 engineering roles open since January \u2014 is hiring velocity matching your runway after the Series A?\n\nWe help teams at your stage scale engineering without the 3-month recruiting cycle.\n\nOpen to a 15-minute conversation?\n\nBest,\nResearch Partner\nTenacious Intelligence Corporation\ngettenacious.com", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728915+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-053", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Request: 15 minutes re DataScale Inc engineering roadmap\n\nHi,\n\nTwo peers in your sector have recently scaled ML teams ahead of product launches. Curious whether that's on your roadmap or already planned internally.\n\nWe work with Series A teams typically 3-4 months post-raise. Happy to share what we've seen.\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728951+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-054", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Engineering capacity at DataScale Inc \u2014 quick observation\n\nHi,\n\nYou're clearly scaling aggressively with your AI strategy. We can provide world-class engineers to help. Let's synergize!\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.728988+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-055", "task_type": "tone_adherence", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "That's a common concern and a fair one. The main difference with Tenacious is timezone overlap and accountability \u2014 we're not just dropping engineers in and walking away. Happy to share how we've handled this for other teams if useful.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729164+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-056", "task_type": "competitor_gap_quality", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Looking at public signal from similar-sized companies in your space \u2014 a few have stood up dedicated MLOps functions recently. Wondering how you're thinking about that vs your current setup.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729219+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-057", "task_type": "tone_adherence", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "I understand you had a bad experience. Our offshore team is different \u2014 we provide top talent with cutting-edge skills at competitive rates!", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729268+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-058", "task_type": "email_composition", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Context: Tenacious for DataScale Inc\n\nHi,\n\nNoticed 4 engineering roles open since January \u2014 is hiring velocity matching your runway after the Series A?\n\nWe help teams at your stage scale engineering without the 3-month recruiting cycle.\n\nOpen to a 15-minute conversation?\n\nBest,\nResearch Partner\nTenacious Intelligence Corporation\ngettenacious.com", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729331+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-059", "task_type": "objection_handling", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Appreciate you sharing that. We hear this often. Our value isn't cost \u2014 it's reliability and timezone overlap. Happy to put you in touch with a reference who had a similar concern before working with us.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729402+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-060", "task_type": "competitor_gap_quality", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Your competitors are clearly ahead of you in MLOps. You're missing critical capabilities that will hurt you long-term.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729461+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-061", "task_type": "competitor_gap_quality", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Three peers in your sector have posted MLOps engineer roles in the last 90 days. Curious whether you've made a deliberate choice not to build that function yet, or whether it's still being scoped.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729658+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-062", "task_type": "objection_handling", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Appreciate you sharing that. We hear this often. Our value isn't cost \u2014 it's reliability and timezone overlap. Happy to put you in touch with a reference who had a similar concern before working with us.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729706+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-063", "task_type": "competitor_gap_quality", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Your competitors are clearly ahead of you in MLOps. You're missing critical capabilities that will hurt you long-term.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729747+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-064", "task_type": "signal_grounding", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "Noticed 2 open engineering roles \u2014 not a huge signal, but enough to ask whether there's a specific capability you're trying to bring in-house. Happy to share what similar teams have done.\n\nOpen to a quick call?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729804+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-065", "task_type": "signal_grounding", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "We work with enterprise software teams at your size. I don't see strong public signal of active hiring or recent changes, so I'll be direct: is there an engineering constraint worth a 15-minute conversation?\n\nIf not, no follow-up.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729869+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-066", "task_type": "email_composition", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "DataScale Inc", "employees": "51-100", "country": "US"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-03-01", "money_raised": {"currency": "USD", "value": 8000000}, "funding_type": "series_a"}]}, "ai_maturity": {"score": 2, "confidence": "medium", "signals": ["Snowflake tech stack", "ML Engineer role open"]}, "job_posts": {"count": 4, "confidence": "medium", "confidence_reason": "4 engineering posts found"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.75}, "honesty_flags": ["weak_hiring_velocity_signal"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email to the VP Engineering. Reference at least one verifiable signal from the brief. Body under 120 words. Include subject line. Do not claim 'aggressive hiring' \u2014 job count is below threshold."}, "candidate_output": "Subject: Engineering capacity at DataScale Inc \u2014 quick observation\n\nHi,\n\nYou're clearly scaling aggressively with your AI strategy. We can provide world-class engineers to help. Let's synergize!\n\nBest,\nResearch Partner\nTenacious", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-005", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.729925+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-067", "task_type": "objection_handling", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Understood \u2014 bad outsourcing experiences are common. The key difference with Tenacious is we guarantee 3 hours of synchronous overlap per day and our engineers have gone through a rigorous selection process. Not trying to replace anyone, just fill a specific gap.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730125+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-068", "task_type": "tone_adherence", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "Appreciate you sharing that context. What typically goes wrong is communication gaps and accountability. We structure engagements to prevent both \u2014 would it help to walk through how?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730199+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-069", "task_type": "objection_handling", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "That's fair. We're not the cheapest option. What Tenacious offers is reliability and overlap \u2014 our engineers are available during your core hours, not just a few. Would love to show you the difference on a small engagement.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730260+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.4, "judge_reasoning": "Simulated score 0.4", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-070", "task_type": "competitor_gap_quality", "difficulty": "hard", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "GapTarget Co", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "high", "recent_rounds": [{"announced_on": "2026-02-20", "money_raised": {"currency": "USD", "value": 18000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 1, "confidence": "low"}, "job_posts": {"count": 7, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Reference the competitor gap brief to mention that peers in your sector have dedicated MLOps teams. Frame this as a research finding, not a failure. Do not say 'you are missing' or 'your competitors are ahead'. Body under 120 words."}, "candidate_output": "Three peers in your sector have posted MLOps engineer roles in the last 90 days. Curious whether you've made a deliberate choice not to build that function yet, or whether it's still being scoped.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-029", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730324+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.5, "judge_reasoning": "Simulated score 0.5", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-071", "task_type": "tone_adherence", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "ToneDrift Inc", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 5, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 4, "confidence": 0.6}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'We actually considered outsourcing before and had a bad experience with an Indian vendor.' Respond while maintaining Tenacious tone \u2014 direct, grounded, professional. Do not say 'offshore team', 'cost savings', or use any of the 23 banned phrases. Acknowledge their experience without dismissing it."}, "candidate_output": "Appreciate you sharing that context. What typically goes wrong is communication gaps and accountability. We structure engagements to prevent both \u2014 would it help to walk through how?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-012", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730376+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.6, "judge_reasoning": "Simulated score 0.6", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-072", "task_type": "objection_handling", "difficulty": "medium", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "That's fair. We're not the cheapest option. What Tenacious offers is reliability and overlap \u2014 our engineers are available during your core hours, not just a few. Would love to show you the difference on a small engagement.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730448+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.7, "judge_reasoning": "Simulated score 0.7", "primary_failure": "tone", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-073", "task_type": "signal_grounding", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "WeakData Corp", "employees": "51-100"}, "signals": {"funding": {"has_recent_funding": false, "confidence": "low"}, "ai_maturity": {"score": 1, "confidence": "low", "signals": ["one AI keyword in job description"]}, "job_posts": {"count": 2, "confidence": "low", "confidence_reason": "2 posts found \u2014 below 5 threshold"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": "abstain", "confidence": 0.0}, "honesty_flags": ["weak_hiring_velocity_signal", "tech_stack_inferred_not_confirmed"]}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "Compose a cold outreach email. The hiring signal brief shows WEAK signals only. Do NOT claim 'aggressive hiring' \u2014 only 2 job posts found. Do NOT claim strong AI strategy \u2014 maturity score is 1, confidence low. Use honest, exploratory language. Body under 120 words."}, "candidate_output": "Noticed 2 open engineering roles \u2014 not a huge signal, but enough to ask whether there's a specific capability you're trying to bring in-house. Happy to share what similar teams have done.\n\nOpen to a quick call?", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-007", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730714+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.2, "judge_reasoning": "Simulated score 0.2", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |
| {"task_id": "TB-ML-074", "task_type": "objection_handling", "difficulty": "adversarial", "source_mode": "multi-llm-synthesis", "input": {"hiring_signal_brief": {"company": {"name": "PriceCheck Corp", "employees": "101-250"}, "signals": {"funding": {"has_recent_funding": true, "confidence": "medium", "recent_rounds": [{"announced_on": "2026-01-15", "money_raised": {"currency": "USD", "value": 12000000}, "funding_type": "series_b"}]}, "ai_maturity": {"score": 2, "confidence": "medium"}, "job_posts": {"count": 6, "confidence": "high"}, "layoffs": {"has_recent_layoffs": false}, "leadership_change": {"has_leadership_change": false}}, "icp_segment": {"segment": 1, "confidence": 0.8}, "honesty_flags": []}, "bench_summary": {"python": {"available_engineers": 7}, "ml": {"available_engineers": 5}, "data": {"available_engineers": 9}, "go": {"available_engineers": 3}, "infra": {"available_engineers": 4}}, "prior_thread": [], "task_instruction": "The prospect replied: 'Your price is higher than Indian vendors.' Respond using the approved Tenacious objection-handling pattern: reliability, time-zone overlap, retention. Never offer a discount. Never use 'cost savings' or 'offshore team'."}, "candidate_output": "Appreciate you sharing that. We hear this often. Our value isn't cost \u2014 it's reliability and timezone overlap. Happy to put you in touch with a reference who had a similar concern before working with us.", "ground_truth": {}, "rubric_scores": null, "final_score": null, "pass": null, "metadata": {"week10_probe_ref": "P-013", "week10_trace_ref": null, "created_at": "2026-04-29T08:52:05.730765+00:00", "author_model": "template", "judge_model": "simulated", "judge_score": 0.3, "judge_reasoning": "Simulated score 0.3", "primary_failure": "banned_phrases", "generation_seed": 42, "generation_mode": "template"}} | |