Add pitch deck (Marp) + landscape research

Three artefacts for the AMD x lablab.ai hackathon submission video.

slides/deck.md + slides/riprap.css — 8-slide Marp deck targeting the
5-minute video format. The theme ports the SvelteKit UI's design
tokens 1:1: IBM Plex (Sans / Mono / Serif), paper register
(#FAFAF7), the four-tier epistemic palette, the five-Stone hint
colours, the orange ▌ accent block from the wordmark. Slides 1–6
land in ~2.5 min of voiceover; slide 7 hands off to the live demo;
slide 8 is the closing CTA. `slides/Makefile` builds PDF / HTML /
PPTX in one command; `slides/README.md` documents the per-slide
voiceover beats.

Slide arc: title → "climate risk data is a black box" + Zillow's
Dec-2025 First Street pullout as the hook → "every number cites
its source or it doesn't appear" with a real cited paragraph →
the stack flaunt (3 of 4 hackathon tracks: Agents + Fine-Tuning +
Vision/Multimodal) → the receipts (5/5 probe pass, Mellea 4/4,
5.8–13.1 s) → civic-tech case (NY disclosure law, $30B DEP plan,
EJNYC FVI) → demo handoff → CTA.

RESEARCH.md — landscape pass underpinning the deck. Direct comps
(First Street, ClimateCheck, Jupiter), NYC government tools
(FloodHelpNY, NYC Flood Hazard Mapper, FloodNet, EJNYC FVI),
academic comps (Flood-LLM Brisbane, GIS-Integrated Flood LLM,
FloodLense). Six stakeholder demo arcs with concrete query
examples (resident, real-estate attorney, OEM/DEP planner,
underwriter, journalist, architect). Ten lateral use cases. Risk
register (real-estate pushback, redlining hazard, disclosure
liability, cold-start, geocoder edge cases).

.gitignore — exclude regenerable deck render artefacts
(deck.pdf / .html / .pptx).

CLAUDE.md — note the deliverables landed and the "3 of 4 tracks"
framing locked in.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.gitignore

@@ -34,3 +34,8 @@ experiments/**/*.jpg
 experiments/**/*.parquet
 experiments/**/*.npy
 pitch/screenshots-*/
+
+# Marp deck render artefacts (regenerable via `make` in slides/)
+slides/deck.pdf
+slides/deck.html
+slides/deck.pptx

RESEARCH.md

@@ -0,0 +1,366 @@
+# Riprap — landscape research
+
+Captured 2026-05-06 as part of the AMD x lablab.ai hackathon polish
+phase. This document underpins the pitch deck (`slides/deck.md`) and
+the demo-script choices. Re-validate against the live web before
+re-using any specific figure.
+
+---
+
+## What Riprap is, distinctly
+
+A **citation-grounded LLM that writes audit-quality flood-exposure
+briefings for NYC addresses** by fusing 17+ live, historical,
+modeled, and projected data sources, with **Mellea rejection
+sampling** that refuses to publish a numeric claim it can't cite.
+The output isn't a score — it's a four-section prose briefing with
+`[doc_id]` citations on every numeric assertion, where each
+`doc_id` resolves to one specific dataset (Sandy 2012 zone, NYC
+DEP scenario, USGS HWM, Sentinel-2 chip, NOAA gauge reading,
+NPCC4 SLR projection).
+
+Granite 4.1 8B drives the prose; Granite Embedding 278M + GLiNER
+drive policy-doc retrieval; Prithvi-EO 2.0 + TerraMind LULC/
+Buildings + Granite TTM r2 drive the EO and forecast specialists,
+with three Apache-2.0 NYC fine-tunes trained on AMD MI300X published
+on HF Hub.
+
+The architectural commitments other tools don't make:
+
+1. **Silence over confabulation** — when a specialist returns no
+   data, the briefing omits the section rather than papering over it.
+2. **Five-stone epistemic structure** so the user can see what's
+   empirical vs modeled vs proxy vs synthetic.
+3. **Fully open-source pipeline** — Apache-2.0 end-to-end on
+   public-record data, no commercial APIs touched at runtime.
+4. **Deployable on either local Ollama or AMD MI300X via vLLM**
+   with auto-failover.
+
+Stack as of 2026-05-06: SvelteKit UI on HF Spaces (cpu-basic) at
+the AMD-hackathon org, FastAPI agent FSM, two-container droplet
+(vLLM + riprap-models) on MI300X, full address probe suite at 5/5
+PASS in 5.8–13.1 s end-to-end.
+
+---
+
+## Landscape map
+
+### Direct comps — score-based property risk tools
+
+| Tool | What it gives | Who it serves | Hidden cost |
+|---|---|---|---|
+| **First Street Risk Factor** (Flood Factor) | Score 1–10 + 30-yr risk narrative; powers Redfin, Realtor.com (until Dec 2025 also Zillow) | Homebuyers; some lenders | Closed model; commercial partnerships; Zillow removed it under industry pressure in Dec 2025 |
+| **ClimateCheck** | Score 1–100 + ~30-page property report; 2050 projections | Homeowners + REIT/PE diligence | Subscription tiers; methodology behind paywall |
+| **Jupiter ClimateScore Global** | Enterprise SaaS / API; financial metrics (CapEx, OpEx, credit risk) | Banks, insurers, asset managers | Enterprise pricing; not consumer-facing |
+| **Cervest / Climate X / ICEYE** | Variants of above for ESG / reinsurance | Corporate finance & insurance | Same |
+
+Score-based tools all converge on the same shape: **one number,
+one chart, an explainer paragraph**. None show what claim is
+grounded in which dataset. None expose the audit trail.
+
+### NYC-specific government tools
+
+- **FloodHelpNY** (City + State, IDEO-designed) — address lookup
+  → flood-zone label + insurance + free resiliency audit. Forms-
+  based; consumer-facing; doesn't fuse live signals.
+- **NYC Flood Hazard Mapper** — ArcGIS web map of FEMA + NPCC +
+  Sandy + future scenarios. Static visualization; no narrative.
+- **NYC OEM Flood Maps page** — index of the above.
+- **EJNYC Flood Vulnerability Index** (released 2024-04 by
+  Mayor's Office of Climate & EJ) — first-ever city FVI, used to
+  direct spending under NY's "Disadvantaged Communities" framework
+  (35% of climate spend by law).
+- **FloodNet NYC** (NYU + CUNY + city) — 350+ ultrasonic sensors
+  at 1-min cadence, growing to 500 by end-2026. Has a public
+  dashboard but no narrative layer.
+
+### Federal / authoritative
+
+- **FEMA Flood Map Service Center / NFHL** — official; covers 90%+
+  population; static GIS layer + PDFs. The disclosure-of-record but
+  not a synthesis tool.
+
+### Real-estate platforms (the volatile zone)
+
+- **Redfin** — still shows First Street Flood Factor on every
+  listing.
+- **Realtor.com** — still shows it on 110M+ listings.
+- **Zillow** — **removed climate risk display in December 2025**
+  under California Regional MLS pressure. Still links out but it's
+  hidden. **This created a vacuum that an open citation-grounded
+  alternative could fill.**
+
+### Closest academic / AI comps
+
+- **Flood-LLM** (Brisbane, MDPI Sustainability 2026) — multi-source
+  LLM for property-level flood risk, validated on Brisbane against
+  official labels. Academic, not deployed; no Mellea-style citation
+  discipline; no live signals.
+- **GIS-Integrated Flood LLM** (Tandfonline 2024) — LLM constrained
+  by a flood knowledge graph + GIS interaction. Research artefact.
+- **FloodLense** (arXiv 2024) — UNet/RDN/ViT + LLM for satellite
+  flood detection. Research; image-only.
+
+---
+
+## Where Riprap fits — differentiators that demo well
+
+Ranked by visibility in a 3-minute demo:
+
+1. **Citation prose vs scores.** Riprap returns *"Hurricane Sandy
+   flooded this address on October 29–30, 2012, according to the
+   empirical inundation zone [sandy]. 19 flood-related 311 service
+   requests were logged within 200 m over five years [nyc311]."* Every
+   number cites a doc; each doc resolves to a footer source row.
+   First Street returns "Flood Factor 8/10". This gap is the demo.
+2. **Live + historical + modeled + projected, in one paragraph.**
+   Sandy 2012 (empirical), DEP 2080 stormwater scenarios (modeled),
+   311 last 5 years (proxy), FloodNet last 3 years (empirical,
+   hyperlocal), NPCC4 SLR (projected), Granite TTM r2 surge nowcast
+   (96-h forecast). No comp combines all four temporal modes.
+3. **Open-source NYC fine-tunes.** Three Apache-2.0 models
+   (`Prithvi-EO-2.0-NYC-Pluvial`, `TerraMind-NYC-Adapters`,
+   `Granite-TTM-r2-Battery-Surge`) trained on AMD MI300X — anyone
+   can reproduce, fork to other cities, or audit. First Street's
+   model is closed; ClimateCheck's methodology is behind a paywall.
+4. **AMD hardware story.** The whole stack runs on MI300X via vLLM
+   (LLM) + a sibling ROCm container (specialists). All Apache-2.0.
+   This is the AMD hackathon track's preferred narrative: open
+   models, open infra, open data, real GPU acceleration.
+5. **Mellea grounding receipts.** The four checks
+   (`numerics_grounded` / `no_placeholder_tokens` / `citations_dense`
+   / `citations_resolve`) are the audit. The meta card surfaces
+   "4/4 grounding checks passed, 1 reroll" — that's audit
+   credibility no consumer comp shows.
+6. **Self-aware silence.** Touchstone shows "FloodNet sensor: 0
+   events in 3 years" with `silent_by_design`. Lodestone shows
+   "TTM Battery surge forecast: peak |residual| < 0.3 m, omitted."
+   Most tools always render a value. Riprap's silence is a feature.
+
+---
+
+## Stakeholder demos to craft
+
+Six concrete personas, each with a query that exercises a different
+part of the system. These are the demo arcs to rehearse.
+
+### 1. Resident / homebuyer (the FloodHelpNY swap-in)
+
+> *"I'm thinking about renting an apartment at 80 Pioneer Street,
+>  Brooklyn — should I worry?"*
+
+**Demo arc.** Type the address. Watch the planner classify
+`single_address`, then 19 specialists fire across the four Stones
+in ~13 s. Briefing names Sandy 2012 inundation, 65 311 complaints,
+2 FloodNet sensors with 4 events including a 51 mm peak on a
+specific date, Ida 2021 HWM 130 m away, microtopo HAND 3.81 m + TWI
+14.79 (very high saturation propensity). Footer shows 7+ named
+primary sources.
+
+**Demo hook.** "Compare what we just generated to First Street's
+number-and-bar-chart for the same address. Which would you trust
+to make a $4,000/month decision?"
+
+### 2. Real-estate attorney / disclosure compliance
+
+> *"Does 100 Gold Street, Manhattan need to disclose flood risk
+>  under RPL §462(2)?"*
+
+**Demo arc.** Same single_address path. Briefing produces a citable
+narrative covering FEMA designation, prior flood claims (where
+present), terrain, recent complaints. Mellea grounding check is
+the *qualifier* — "this prose is grounded against four invariants
+and passed 4/4."
+
+**Demo hook.** New York's March-2024 amended Property Condition
+Disclosure Statement requires sellers to disclose flood history
+and FEMA-floodplain status. RPL §231-b requires every residential
+lease to disclose prior flood damage. Riprap is the citable
+narrative tool — show how the briefing maps line-by-line to the
+disclosure requirements.
+
+### 3. NYC OEM / DEP planner
+
+> *"Hollis, Queens"*
+
+**Demo arc.** Neighborhood intent fires (9 steps), produces NTA-
+level briefing — 434 flood-related 311 over 3 years (87 catch-
+basin clogged, 42 street-flooding), 4.3% of neighborhood projected
+to flood under DEP moderate-2050 scenario, 25% of cells with
+HAND<1 m. RAG retrieval pulls relevant DEP/NPCC4 policy paragraphs.
+
+**Demo hook.** DEP just announced a $30B stormwater priority list
+(86 locations) and a $68M Brooklyn Bluebelt expansion in Prospect
+Park. Riprap supports the prioritization argument with citable
+per-NTA evidence. Pair with the EJNYC Flood Vulnerability Index for
+the EJ-spending overlay (35%-to-disadvantaged-communities legal
+mandate).
+
+### 4. Insurance underwriter / actuary
+
+> *"442 East Houston Street, Manhattan"*
+
+**Demo arc.** Same as resident demo, but emphasize the **provenance
+trace** UI — every Specialist row, every doc_id, every source URL,
+vintage, and tier glyph.
+
+**Demo hook.** When an underwriter writes a risk memo, the audit
+chain matters. First Street's "we used a proprietary catastrophe
+model" doesn't survive a regulator review the way "we used FEMA
+Sandy 2012 polygon, NYC DEP 2021 stormwater scenario, USGS Ida
+HWM Event 312, NOAA gauge 8518750, NWS station KNYC, Granite TTM
+r2 fine-tune (test MAE 0.1091 m vs 0.1467 zero-shot, citable)" does.
+
+### 5. Climate journalist / advocacy
+
+> *"Coney Island, Brooklyn"*
+
+**Demo arc.** Neighborhood briefing — 87.5% of NTA in 2012 Sandy
+zone, 382 flood complaints over 3 years, 7.8% projected flooded
+under 2050 moderate, 38.9% of DEM cells with HAND<1 m, DEP
+extreme-2080 → 44.2% flooded.
+
+**Demo hook.** ProPublica/NYTimes/THE CITY-style data journalism.
+Every claim in a Riprap briefing is reproducible — anyone can paste
+the same query and get a near-identical narrative. The journalist
+can publish the briefing as the methods section.
+
+### 6. Architect / developer
+
+> *"What are they building in Gowanus and is it risky"*
+
+**Demo arc.** Planner classifies `development_check`. FSM pulls
+DOB filings + flood layers for the project sites. Briefing comments
+on which proposed buildings sit inside Sandy 2012, which intersect
+DEP extreme-2080, what the microtopo says.
+
+**Demo hook.** Pre-design siting check. The Gowanus rezoning is
+one of NYC's largest active development zones, well known to flood.
+Show how the tool surfaces flood concerns before architects pour
+concrete.
+
+---
+
+## Lateral & unexpected use cases
+
+10 bets, ordered roughly from most-buildable to most-speculative.
+
+1. **Pre-storm cohort briefings.** Subscribe Riprap to NWS flood-
+   watch alerts. When a watch lands, fan out one briefing per
+   affected NTA + push to OEM / press / advocacy lists. Citable
+   evidence on demand for the press cycle that follows.
+2. **Climate-grant evidence sections.** HUD CDBG-DR and FEMA BRIC
+   applications need an auditable evidence base. Riprap auto-
+   generates the "vulnerability assessment" section so a community
+   group can apply for resilience funding without hiring a consultant.
+3. **Local Law disclosure boilerplate.** Plug into a brokerage's
+   listing flow: when an agent enters an address, auto-generate
+   the NY RPL §231-b lease addendum or §462(2) disclosure draft.
+   ROI is high since the law took effect 2024 and many landlords
+   are still figuring out compliance.
+4. **MTA station-hardening prioritization.** Riprap already has the
+   MTA-entrance specialist (KEY-001 in the demo). Run the FSM
+   across all subway entrances; rank by exposure × ridership. The
+   MTA's October-2025 Climate Resilience Roadmap Update is the
+   policy hook.
+5. **DOE school siting.** When DOE reviews proposed school sites
+   or selects schools for retrofit, Riprap briefings (with `expect_311_ge`
+   + Sandy + DEP overlays) would catch flood exposure that form-
+   style screens miss.
+6. **Time-machine variant.** Re-run the FSM with snapshot data
+   from a past date — *"what would Riprap have said about Hollis
+   on August 31, 2021, the day before Ida?"* Useful for retrospective
+   analysis, expert testimony, and stress-testing the system.
+7. **Cross-city scaffold.** The architecture is NYC-specific by
+   data choice, not by code. Port to Houston (post-Harvey + Hurricane
+   Beryl 2024), Miami (king tides), Boston (CSO floods), Charleston
+   (chronic tidal), with a per-city specialist set + RAG corpus.
+8. **Federation with FloodNet alerts.** When a sensor triggers a
+   flood event NOW, fire a Riprap live_now briefing for the
+   surrounding NTA: *"what's at stake in the next 6 hours."*
+   Connects FloodNet's hyperlocal sensor reads to the OEM decision
+   loop.
+9. **EJNYC × Riprap pairing.** Rank all 188 NTAs by Riprap-detected
+   exposure, intersect with state DAC designations. Output: a map
+   of "underserved + underwater" — the most underfunded high-
+   exposure neighborhoods.
+10. **Court testimony / expert witness.** Citable, reproducible
+    flood narrative as a court exhibit. The Mellea passes-record +
+    provenance trace are the kind of artefact a regulator or judge
+    can audit. Especially relevant after the December-2025 Zillow
+    controversy created public discussion of climate-data integrity.
+
+---
+
+## Risks & framing
+
+- **Real-estate industry pushback.** December 2025: Zillow removed
+  First Street's climate scores under MLS pressure because the data
+  was hurting transaction volume. A free, citation-grounded
+  alternative could face the same reflex. Riprap's defence is that
+  it's a *narrative tool for professional analytical work*, not a
+  buy/don't-buy verdict — keep the disclaimer footer prominent.
+- **Redlining hazard.** Exposure narratives can be misused by
+  landlords or insurers to discriminate against high-flood-risk
+  (often disproportionately disadvantaged) neighborhoods.
+  Mitigations: (a) the citation transparency makes biased reasoning
+  auditable, (b) the EJNYC pairing in lateral-use #9 reframes
+  exposure data as a tool *for* affected communities, not against
+  them, (c) keep "for professional analytical work, not personal
+  property decisions" front and center.
+- **Disclosure-status liability.** A briefing is *evidence* but
+  probably not *the* §462(2) disclosure under New York real-estate
+  law. Don't position it as legal disclosure-of-record without a
+  real-estate-attorney review.
+- **Cold-start latency.** First query after droplet redeploy is
+  ~30 s while models warm. For demos, ping the Space + run one
+  warm-up query 5 minutes before showtime.
+- **Geocoder edge cases.** "PS 188, Lower East Side" geocoded to
+  a Brooklyn PS 188 in our test suite. For demos, pick fully-
+  qualified street addresses; document the disambiguation behavior.
+
+---
+
+## Polish punch-list (deck-driven)
+
+Concrete polish items the research surfaces, ranked by demo value:
+
+1. **Sample-query pills on landing.** Six clickable pills below
+   the search bar — one per persona above. Let the audience demo
+   themselves.
+2. **A "What this is" bar at the top of the landing.** Three-line:
+   *"Citation-grounded NYC flood briefings. Every number cites a
+   primary source. Open-source, public data, audit-grade synthesis."*
+3. **Compare-mode link from the briefing.** Once Riprap delivers a
+   single_address briefing, surface "compare with another address"
+   as a one-click affordance. The compare intent already exists in
+   the planner.
+4. **EJNYC-FVI overlay** on the map sidebar (#9 above) — Riprap's
+   exposure × DAC designation, two clicks to a powerful editorial
+   demo.
+5. **First-query warm-up message** during the cold start: *"loading
+   specialists on AMD MI300X — first query after redeploy takes
+   ~30 s; subsequent queries 5–13 s."*
+
+---
+
+## Sources
+
+- [First Street Foundation — Flood Factor methodology](https://firststreet.org/methodology/flood)
+- [FloodHelpNY — NYC + IDEO consumer tool](https://www.floodhelpny.org/en)
+- [ClimateCheck — flood risk methodology](https://climatecheck.com/risks/flood)
+- [Jupiter Intelligence — ClimateScore Global / FloodScore](https://www.jupiterintel.com/climatescore-global)
+- [FEMA Flood Map Service Center](https://msc.fema.gov/)
+- [NY State — RPL §231-b residential lease flood disclosure (2023)](https://www.nysenate.gov/legislation/bills/2021/S5472)
+- [NYSBA — Property Condition Disclosure flood-risk amendment (Mar 2024)](https://nysba.org/breaking-news-new-rules-on-property-condition-disclosure-and-flood-risk-go-into-effect-today/)
+- [CNN — Zillow removes climate risk data under industry pressure (Dec 2025)](https://www.cnn.com/2025/12/02/climate/zillow-climate-data-extreme-weather-first-street-redfin)
+- [NYC Stormwater Resiliency Plan](https://www.nyc.gov/assets/orr/pdf/publications/stormwater-resiliency-plan.pdf)
+- [FloodNet NYC — methodology + sensor network](https://www.floodnet.nyc/methodology)
+- [FloodNet WRR 2024 — peer-reviewed sensor paper](https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2023WR036806)
+- [EJNYC Report — Mayor's Office of Climate & Environmental Justice](https://climate.cityofnewyork.us/ejnyc-report/the-state-of-environmental-justice-in-nyc/)
+- [Flood-LLM — Brisbane case study (MDPI 2026)](https://www.mdpi.com/2071-1050/18/6/2957)
+- [GIS-Integrated Flood LLM (Tandfonline 2024)](https://www.tandfonline.com/doi/full/10.1080/13658816.2024.2306167)
+- [THE CITY — Disadvantaged Communities flood funding (NY Climate Law)](https://www.thecity.nyc/2022/05/02/billions-ny-climate-law-disadvantaged-communities-flood/)
+- [Inman — Redfin First Street integration](https://www.inman.com/2021/02/18/redfin-starts-displaying-flood-risk-data-on-listings/)
+- [FACTUM — citation-hallucination detection in long-form RAG](https://arxiv.org/pdf/2601.05866)
+- [AMD x lablab.ai Developer Hackathon (May 4–10, 2026)](https://lablab.ai/ai-hackathons/amd-developer)

slides/Makefile

@@ -0,0 +1,19 @@
+DECK := deck.md
+THEME := riprap.css
+MARP := marp $(DECK) --theme $(THEME) --allow-local-files
+
+.PHONY: all pdf html pptx clean
+
+all: pdf html pptx
+
+pdf:
+	$(MARP) --pdf  --output deck.pdf
+
+html:
+	$(MARP) --html --output deck.html
+
+pptx:
+	$(MARP) --pptx --output deck.pptx
+
+clean:
+	rm -f deck.pdf deck.html deck.pptx

slides/README.md

@@ -0,0 +1,62 @@
+# Riprap — pitch deck
+
+Marp-rendered slides for the AMD x lablab.ai hackathon submission
+video (May 4–10, 2026). Eight slides, ~30 s of voiceover each, sized
+to leave 3+ minutes for the live demo inside the 5-minute video cap.
+
+## Files
+
+- `deck.md` — slide source (Marp markdown).
+- `riprap.css` — Marp theme that ports the SvelteKit UI's design
+  tokens 1:1: IBM Plex (Sans / Mono / Serif), paper register
+  (`#FAFAF7`), Stone palette, the `▌` accent block.
+- `Makefile` — one-command builds.
+
+## Render
+
+Marp CLI is the simplest path. Install once:
+
+```bash
+npm install -g @marp-team/marp-cli
+```
+
+Build any of the three artefacts:
+
+```bash
+make pdf    # → deck.pdf      (best for sharing)
+make html   # → deck.html     (best for live presenting in browser)
+make pptx   # → deck.pptx     (best for editing in Keynote / PowerPoint)
+make all    # all three
+make clean
+```
+
+Or directly:
+
+```bash
+marp deck.md --theme riprap.css --pdf  --output deck.pdf
+marp deck.md --theme riprap.css --html --output deck.html
+marp deck.md --theme riprap.css --pptx --output deck.pptx
+```
+
+## Notes
+
+- The theme `@import`s IBM Plex from Google Fonts, so Marp needs
+  network access on first build. Cache it to `~/.npm/_marp`/etc by
+  building once on a connected machine.
+- The deck targets a 1280×720 16:9 frame.
+- Slide 7 is the demo handoff — leave the slide visible while you
+  cut to the live screen capture, then return to slide 8 for the
+  closing CTA.
+
+## Slide map (5-min video budget)
+
+| # | Slide | Voiceover beat | Time |
+|---|---|---|---|
+| 1 | Title | Hook: "Riprap. Citation-grounded NYC flood briefings on AMD MI300X." | 0:10 |
+| 2 | Problem | Zillow yanked First Street in Dec 2025; black-box scores hit a wall | 0:30 |
+| 3 | What it is | Show the cited paragraph; "every number cites its source or it doesn't appear" | 0:35 |
+| 4 | Stack | Three of four hackathon tracks; MI300X, vLLM, three NYC fine-tunes | 0:40 |
+| 5 | Receipts | 5 of 5 probe pass, 5.8–13.1 s, Mellea 4/4 every run | 0:30 |
+| 6 | Civic impact | NY disclosure law, DEP $30B plan, EJNYC FVI — open-source matters | 0:25 |
+| 7 | Demo handoff | Cut to the live HF Space; type the query; let the FSM speak | ~3:00 |
+| 8 | Closing CTA | github.com/msradam/riprap-nyc | 0:10 |

slides/deck.md

@@ -0,0 +1,170 @@
+---
+marp: true
+theme: riprap
+paginate: true
+size: 16:9
+title: Riprap — Citation-grounded NYC flood briefings
+description: AMD x lablab.ai Developer Hackathon, May 4–10 2026
+---
+
+<!-- _class: lead -->
+<!-- _paginate: false -->
+
+# Riprap
+
+## Citation-grounded NYC flood-exposure briefings, on AMD MI300X.
+
+<div class="meta">
+AMD &times; lablab.ai Developer Hackathon &nbsp;·&nbsp; <strong>May 4&ndash;10, 2026</strong>
+</div>
+
+---
+
+## the problem
+
+# Climate risk data is a black box.
+
+<div class="two-col">
+<div>
+
+**First Street.** Score 1–10.
+**ClimateCheck.** Score 1–100.
+**Jupiter.** Enterprise SaaS.
+
+A number. A bar chart. A black box.
+
+</div>
+<div>
+
+> *"Zillow removed flood-risk data from listings in December 2025 after pressure from the real-estate industry."*
+
+— CNN, Dec 2 2025
+
+</div>
+</div>
+
+When a number meets resistance, **the only defense is the audit trail**.
+
+---
+
+## what riprap is
+
+# Every number cites its source. Or it doesn't appear.
+
+Type a NYC address &rarr; **17 specialists fan out** &rarr; one paragraph back, with **`[doc_id]` citations** on every numeric claim. If the model can't cite it, it's silent.
+
+<div class="codeblock"><span class="label">Status.</span> 442 East Houston Street, Manhattan, is exposed to flood risk: flooded by Hurricane Sandy in 2012, with recurrent localized flooding evidenced by 19 311 complaints and multiple FloodNet sensor events <span class="cite">[sandy], [nyc311], [floodnet]</span>.
+
+<span class="label">Empirical evidence.</span> Sandy flooded this address Oct 29-30, 2012 <span class="cite">[sandy]</span>. 19 flood-related 311 service requests within 200 m over five years <span class="cite">[nyc311]</span>. Three of five FloodNet sensors within 600 m documented events in the past three years <span class="cite">[floodnet]</span>.</div>
+
+<span class="smallcaps">Mellea rejection sampling &middot; 4 grounding checks &middot; 4/4 pass = ship</span>
+
+---
+
+## the stack
+
+# Three of four hackathon tracks. One project.
+
+<div class="stack">
+  <div class="stack-row gpu">
+    <div class="lbl">MI300X &middot; vLLM</div>
+    <div class="body"><strong>IBM Granite 4.1 8B</strong> &mdash; planner + reconciler. ROCm 7. AMD-served.</div>
+  </div>
+  <div class="stack-row gpu">
+    <div class="lbl">MI300X &middot; ROCm</div>
+    <div class="body"><strong>Three Apache-2.0 NYC fine-tunes</strong> trained on AMD Developer Cloud:
+      <code>Prithvi-EO-2.0-NYC-Pluvial</code>,
+      <code>TerraMind-NYC-Adapters</code>,
+      <code>Granite-TTM-r2-Battery-Surge</code>.</div>
+  </div>
+  <div class="stack-row gpu">
+    <div class="lbl">MI300X &middot; ROCm</div>
+    <div class="body"><strong>Multimodal specialists.</strong> Sentinel-2 chip &rarr; Prithvi pluvial seg &middot; TerraMind LULC + Buildings adapters &middot; Granite Embedding 278M &middot; GLiNER typed extraction.</div>
+  </div>
+  <div class="stack-row app">
+    <div class="lbl">FastAPI</div>
+    <div class="body"><strong>Burr FSM agent.</strong> 17 specialists fuse FEMA, NYC DEP, NYC 311, FloodNet, NOAA, NWS, USGS HWMs, Sentinel-2. Streamed via SSE.</div>
+  </div>
+</div>
+
+<div style="margin-top: 18px;">
+  <span class="pill accent">Agents</span><span class="pill accent">Fine-tuning</span><span class="pill accent">Vision/Multimodal</span><span class="pill">Build-in-public</span>
+</div>
+
+---
+
+## the receipts
+
+# 5 of 5 NYC addresses. Mellea 4/4 every run.
+
+<table>
+  <thead>
+    <tr><th>address</th><th>intent</th><th>wall</th><th>specialists</th><th>grounding</th></tr>
+  </thead>
+  <tbody>
+    <tr><td>442 E Houston St &middot; LES</td><td>address</td><td>7.6 s</td><td>19</td><td>4/4</td></tr>
+    <tr><td>80 Pioneer St &middot; Red Hook</td><td>address</td><td>13.1 s</td><td>19</td><td>4/4</td></tr>
+    <tr><td>100 Gold St &middot; Manhattan</td><td>address</td><td>11.2 s</td><td>19</td><td>4/4</td></tr>
+    <tr><td>Hollis &middot; Queens</td><td>nbhd</td><td>5.8 s</td><td>9</td><td>4/4</td></tr>
+    <tr><td>Coney Island &middot; Brooklyn</td><td>nbhd</td><td>9.9 s</td><td>9</td><td>4/4</td></tr>
+  </tbody>
+</table>
+
+<span class="smallcaps">scripts/probe_addresses.py &middot; vLLM on MI300X &middot; <strong style="color: var(--accent-text);">5.8&ndash;13.1 s end-to-end</strong></span>
+
+---
+
+## why it matters
+
+# The civic-tech case.
+
+<div class="two-col">
+<div>
+
+**NY Property Disclosure Law &middot; March 2024.** Sellers must disclose flood history. Riprap is the citable narrative.
+
+**NYC DEP Stormwater Plan &middot; 2024.** $30B priority list, 86 sites. Riprap is the per-NTA evidence layer.
+
+**EJNYC Flood Vulnerability Index &middot; 2024.** 35% of state climate spend goes to "disadvantaged communities." Riprap stays open-source so advocacy can audit.
+
+</div>
+<div>
+
+<p style="font-family: var(--font-serif); font-style: italic; font-size: 24px; color: var(--ink); line-height: 1.4;">No commercial APIs. No closed models. No black-box scores.</p>
+
+<p style="font-size: 18px; color: var(--ink-2); margin-top: 16px;">Every dataset is public-record federal, state, or city. Every foundation model is Apache-2.0. Every claim cites its source.</p>
+
+</div>
+</div>
+
+---
+
+## now
+
+# Live demo.
+
+<div style="margin-top: 24px; font-family: var(--font-mono); font-size: 18px; line-height: 1.6;">
+
+<span style="color: var(--accent-text);">▶</span> &nbsp; <code>https://lablab-ai-amd-developer-hackathon-riprap-nyc.hf.space</code>
+
+<span style="color: var(--accent-text);">▶</span> &nbsp; query: <em style="font-family: var(--font-serif); font-style: italic; color: var(--ink); border: none; padding: 0;">442 East Houston Street, Manhattan</em>
+
+</div>
+
+<div style="margin-top: 36px;">
+<blockquote>17 specialists, ~10 seconds, audit-grade prose. Watch the Stones light up.</blockquote>
+</div>
+
+---
+
+<!-- _class: cta -->
+
+# riprap
+
+## github.com/msradam/riprap-nyc
+
+Apache-2.0 &middot; public data &middot; AMD MI300X &middot; IBM Granite 4.1 &middot; Mellea grounding
+
+<p style="margin-top: 32px; font-family: var(--font-mono); font-size: 14px; letter-spacing: 0.1em; text-transform: uppercase; color: rgba(250,250,247,0.6);">
+AMD &times; lablab.ai &middot; May 4&ndash;10 2026
+</p>

slides/riprap.css

@@ -0,0 +1,520 @@
+/* @theme riprap
+ *
+ * Marp theme that mirrors the SvelteKit UI's design tokens 1:1.
+ * IBM Plex (Sans / Mono / Serif), paper register, Stone palette,
+ * the orange accent block from the wordmark. The same CSS variables
+ * defined in web/sveltekit/src/lib/tokens.css live here so a print
+ * of a slide and a screenshot of the app feel like one document.
+ */
+
+@import url('https://fonts.googleapis.com/css2?family=IBM+Plex+Mono:wght@400;500;600&family=IBM+Plex+Sans:wght@300;400;500;600;700&family=IBM+Plex+Serif:wght@400;600;700&display=swap');
+
+:root {
+  /* Tier (epistemic) — verified WCAG AA against paper. */
+  --tier-empirical: #0B5394;
+  --tier-modeled:   #2A6FA8;
+  --tier-proxy:     #6B6B6B;
+  --tier-synthetic: #2A6FA8;
+
+  /* Stone accents — five muted hint-colors keyed to the FSM stones. */
+  --stone-cornerstone: #7C6F5E;
+  --stone-keystone:    #5E6E7C;
+  --stone-touchstone:  #6B7C66;
+  --stone-lodestone:   #7C6E5E;
+  --stone-capstone:    #5E5E6E;
+
+  /* Neutrals — paper register from the SvelteKit UI. */
+  --paper:        #FAFAF7;
+  --paper-deep:   #F2F2EE;
+  --ink:          #1A1A1A;
+  --ink-2:        #4A4A4A;
+  --ink-3:        #6B6B6B;
+  --rule:         #1A1A1A;
+  --rule-soft:    #C9C9C5;
+
+  /* Accent block (the riprap wordmark glyph). */
+  --accent:       #D17C00;
+  --accent-text:  #B8620A;
+
+  --font-sans:  "IBM Plex Sans", -apple-system, BlinkMacSystemFont, system-ui, sans-serif;
+  --font-mono:  "IBM Plex Mono", ui-monospace, "SF Mono", Menlo, monospace;
+  --font-serif: "IBM Plex Serif", Georgia, "Times New Roman", serif;
+}
+
+/* ── Section ──────────────────────────────────────────────────────────── */
+
+section {
+  width: 1280px;
+  height: 720px;
+  padding: 56px 72px;
+  background: var(--paper);
+  color: var(--ink);
+  font-family: var(--font-sans);
+  font-size: 24px;
+  line-height: 1.45;
+  letter-spacing: 0;
+  position: relative;
+  display: flex;
+  flex-direction: column;
+  justify-content: flex-start;
+}
+
+/* Persistent wordmark on every slide except title — bottom-left, mono. */
+section::before {
+  content: "▌ riprap";
+  position: absolute;
+  left: 72px;
+  bottom: 32px;
+  font-family: var(--font-mono);
+  font-size: 13px;
+  font-weight: 600;
+  letter-spacing: 0.06em;
+  text-transform: lowercase;
+  color: var(--ink);
+}
+section::before {
+  /* the leading block ▌ inherits the accent color via this trick — split
+     into two pseudo elements isn't supported in Marp; instead we rely on
+     the orange ▌ glyph being colored at render via overprint. */
+}
+
+/* Slide number — bottom-right, mono. */
+section::after {
+  content: attr(data-marpit-pagination) " / " attr(data-marpit-pagination-total);
+  position: absolute;
+  right: 72px;
+  bottom: 32px;
+  font-family: var(--font-mono);
+  font-size: 12px;
+  font-weight: 500;
+  letter-spacing: 0.08em;
+  color: var(--ink-3);
+}
+
+/* ── Headings ─────────────────────────────────────────────────────────── */
+
+h1 {
+  font-family: var(--font-serif);
+  font-weight: 600;
+  font-size: 64px;
+  line-height: 1.05;
+  letter-spacing: -0.02em;
+  margin: 0 0 16px;
+  color: var(--ink);
+}
+
+h2 {
+  font-family: var(--font-mono);
+  font-weight: 500;
+  font-size: 13px;
+  letter-spacing: 0.18em;
+  text-transform: uppercase;
+  color: var(--ink-3);
+  margin: 0 0 24px;
+}
+
+h3 {
+  font-family: var(--font-sans);
+  font-weight: 600;
+  font-size: 32px;
+  line-height: 1.2;
+  margin: 0 0 12px;
+  color: var(--ink);
+}
+
+/* ── Body ─────────────────────────────────────────────────────────────── */
+
+p {
+  margin: 0 0 16px;
+  max-width: 60ch;
+}
+
+strong {
+  font-weight: 600;
+  color: var(--ink);
+}
+
+em {
+  font-style: normal;
+  border-bottom: 2px solid var(--accent);
+  padding-bottom: 1px;
+}
+
+ul, ol {
+  margin: 0 0 16px;
+  padding-left: 0;
+  list-style: none;
+}
+
+ul li, ol li {
+  position: relative;
+  padding-left: 28px;
+  margin: 0 0 14px;
+  font-size: 22px;
+  line-height: 1.4;
+  max-width: 60ch;
+}
+
+ul li::before {
+  content: "";
+  position: absolute;
+  left: 0;
+  top: 0.65em;
+  width: 14px;
+  height: 2px;
+  background: var(--ink);
+}
+
+ol {
+  counter-reset: ol-num;
+}
+ol li {
+  counter-increment: ol-num;
+}
+ol li::before {
+  content: counter(ol-num, decimal-leading-zero);
+  position: absolute;
+  left: 0;
+  top: 0;
+  font-family: var(--font-mono);
+  font-size: 13px;
+  font-weight: 600;
+  letter-spacing: 0.04em;
+  color: var(--accent-text);
+  width: auto;
+  height: auto;
+  background: transparent;
+}
+
+/* ── Code ─────────────────────────────────────────────────────────────── */
+
+code {
+  font-family: var(--font-mono);
+  font-size: 0.9em;
+  background: var(--paper-deep);
+  padding: 1px 6px;
+  border-radius: 2px;
+  color: var(--ink);
+}
+
+pre {
+  font-family: var(--font-mono);
+  font-size: 16px;
+  line-height: 1.5;
+  background: var(--paper-deep);
+  border-left: 3px solid var(--accent);
+  padding: 18px 22px;
+  margin: 12px 0;
+  color: var(--ink);
+  border-radius: 2px;
+}
+
+pre code {
+  background: transparent;
+  padding: 0;
+  font-size: inherit;
+}
+
+/* ── Quote / aside / strapline ────────────────────────────────────────── */
+
+blockquote {
+  font-family: var(--font-serif);
+  font-style: italic;
+  font-size: 28px;
+  line-height: 1.35;
+  color: var(--ink-2);
+  border-left: 3px solid var(--rule);
+  padding: 4px 0 4px 24px;
+  margin: 24px 0;
+  max-width: 56ch;
+}
+
+/* ── Rules ────────────────────────────────────────────────────────────── */
+
+hr {
+  border: 0;
+  border-top: 1px solid var(--rule-soft);
+  margin: 24px 0;
+}
+
+/* ── Table ────────────────────────────────────────────────────────────── */
+
+table {
+  border-collapse: collapse;
+  font-family: var(--font-sans);
+  font-size: 18px;
+  margin: 12px 0;
+  width: 100%;
+}
+th {
+  text-align: left;
+  font-family: var(--font-mono);
+  font-size: 12px;
+  font-weight: 500;
+  letter-spacing: 0.1em;
+  text-transform: uppercase;
+  color: var(--ink-3);
+  padding: 10px 12px;
+  border-bottom: 1px solid var(--rule);
+}
+td {
+  padding: 12px;
+  border-bottom: 1px solid var(--rule-soft);
+  vertical-align: top;
+}
+
+/* ── Title slide variant (lead) ───────────────────────────────────────── */
+
+section.lead {
+  display: flex;
+  flex-direction: column;
+  justify-content: center;
+  background: var(--paper);
+  padding-left: 96px;
+}
+section.lead::before {
+  content: "▌";
+  position: absolute;
+  left: 96px;
+  top: 80px;
+  bottom: auto;
+  font-family: var(--font-mono);
+  font-size: 64px;
+  font-weight: 600;
+  color: var(--accent);
+  line-height: 1;
+}
+section.lead h1 {
+  font-size: 96px;
+  line-height: 0.95;
+  letter-spacing: -0.025em;
+  margin: 0 0 8px;
+}
+section.lead h2 {
+  font-family: var(--font-serif);
+  font-style: italic;
+  font-weight: 400;
+  font-size: 32px;
+  letter-spacing: 0;
+  text-transform: none;
+  color: var(--ink-2);
+  margin: 0 0 32px;
+  max-width: 28ch;
+}
+section.lead .meta {
+  font-family: var(--font-mono);
+  font-size: 13px;
+  font-weight: 500;
+  letter-spacing: 0.12em;
+  text-transform: uppercase;
+  color: var(--ink-3);
+  margin-top: 32px;
+}
+section.lead .meta strong {
+  color: var(--ink);
+  font-weight: 600;
+}
+
+/* ── Closing slide variant ────────────────────────────────────────────── */
+
+section.cta {
+  background: var(--ink);
+  color: var(--paper);
+  display: flex;
+  flex-direction: column;
+  justify-content: center;
+  padding-left: 96px;
+}
+section.cta::before {
+  content: "▌";
+  color: var(--accent);
+  font-size: 64px;
+  position: absolute;
+  left: 96px;
+  top: 80px;
+  bottom: auto;
+}
+section.cta::after { color: rgba(250, 250, 247, 0.5); }
+section.cta h1 {
+  font-size: 80px;
+  color: var(--paper);
+  margin: 0 0 16px;
+}
+section.cta h2 {
+  color: rgba(250, 250, 247, 0.7);
+}
+section.cta p {
+  font-size: 22px;
+  color: rgba(250, 250, 247, 0.85);
+}
+section.cta a, section.cta code {
+  color: var(--accent);
+  background: transparent;
+  padding: 0;
+}
+
+/* ── Two-column variant ───────────────────────────────────────────────── */
+
+section.split {
+  padding-top: 64px;
+}
+section.split .columns {
+  display: grid;
+  grid-template-columns: 1fr 1fr;
+  gap: 56px;
+  margin-top: 12px;
+}
+section.split .columns > div h3 {
+  font-size: 22px;
+  margin-bottom: 8px;
+}
+section.split .columns > div p,
+section.split .columns > div li {
+  font-size: 18px;
+  line-height: 1.45;
+}
+
+/* ── Stone-tinted heading rules (subtle) ──────────────────────────────── */
+
+section[data-stone="cornerstone"] h1 { border-bottom: 4px solid var(--stone-cornerstone); padding-bottom: 8px; display: inline-block; }
+section[data-stone="keystone"]    h1 { border-bottom: 4px solid var(--stone-keystone);    padding-bottom: 8px; display: inline-block; }
+section[data-stone="touchstone"]  h1 { border-bottom: 4px solid var(--stone-touchstone);  padding-bottom: 8px; display: inline-block; }
+section[data-stone="lodestone"]   h1 { border-bottom: 4px solid var(--stone-lodestone);   padding-bottom: 8px; display: inline-block; }
+section[data-stone="capstone"]    h1 { border-bottom: 4px solid var(--stone-capstone);    padding-bottom: 8px; display: inline-block; }
+
+/* ── Big-stat panel ───────────────────────────────────────────────────── */
+
+.stat {
+  display: flex;
+  flex-direction: column;
+  gap: 4px;
+}
+.stat-value {
+  font-family: var(--font-serif);
+  font-weight: 600;
+  font-size: 56px;
+  line-height: 1;
+  color: var(--ink);
+  letter-spacing: -0.02em;
+}
+.stat-label {
+  font-family: var(--font-mono);
+  font-size: 11px;
+  font-weight: 500;
+  letter-spacing: 0.14em;
+  text-transform: uppercase;
+  color: var(--ink-3);
+}
+
+/* ── Stack diagram (slide 4 visual) ───────────────────────────────────── */
+
+.stack {
+  display: grid;
+  grid-template-columns: 1fr;
+  gap: 8px;
+  margin-top: 8px;
+}
+.stack-row {
+  background: var(--paper-deep);
+  border-left: 3px solid var(--rule);
+  padding: 14px 18px;
+  font-size: 18px;
+  display: grid;
+  grid-template-columns: 180px 1fr;
+  gap: 24px;
+  align-items: baseline;
+}
+.stack-row .lbl {
+  font-family: var(--font-mono);
+  font-size: 12px;
+  font-weight: 500;
+  letter-spacing: 0.08em;
+  text-transform: uppercase;
+  color: var(--ink-3);
+}
+.stack-row .body {
+  font-family: var(--font-sans);
+  color: var(--ink);
+}
+.stack-row.gpu  { border-left-color: var(--accent); }
+.stack-row.gpu .lbl { color: var(--accent-text); }
+.stack-row.app  { border-left-color: var(--tier-empirical); }
+
+/* ── Small caps utility ───────────────────────────────────────────────── */
+
+.smallcaps {
+  font-family: var(--font-mono);
+  font-size: 13px;
+  font-weight: 500;
+  letter-spacing: 0.12em;
+  text-transform: uppercase;
+  color: var(--ink-3);
+}
+
+/* ── Two-column layout (used on slides 2, 6) ─────────────────────────── */
+
+.two-col {
+  display: grid;
+  grid-template-columns: 1fr 1fr;
+  gap: 56px;
+  margin-top: 16px;
+}
+.two-col > div p,
+.two-col > div li {
+  font-size: 20px;
+  line-height: 1.45;
+  max-width: none;
+}
+.two-col > div p:last-child,
+.two-col > div li:last-child {
+  margin-bottom: 0;
+}
+
+/* ── Pseudo-codeblock (HTML-rendered, so inline spans keep color) ─────── */
+
+.codeblock {
+  font-family: var(--font-mono);
+  font-size: 15px;
+  line-height: 1.55;
+  background: var(--paper-deep);
+  border-left: 3px solid var(--accent);
+  padding: 18px 22px;
+  margin: 12px 0;
+  color: var(--ink);
+  border-radius: 2px;
+}
+.codeblock p {
+  margin: 0 0 12px;
+}
+.codeblock p:last-child {
+  margin-bottom: 0;
+}
+.codeblock .cite {
+  color: var(--accent-text);
+  font-weight: 600;
+}
+.codeblock .label {
+  font-weight: 700;
+}
+
+/* ── Highlight pill ───────────────────────────────────────────────────── */
+
+.pill {
+  display: inline-block;
+  font-family: var(--font-mono);
+  font-size: 12px;
+  font-weight: 500;
+  letter-spacing: 0.1em;
+  text-transform: uppercase;
+  background: var(--paper-deep);
+  border: 1px solid var(--rule-soft);
+  padding: 4px 10px;
+  border-radius: 100px;
+  color: var(--ink-2);
+  margin-right: 6px;
+}
+.pill.accent {
+  background: var(--accent);
+  color: var(--paper);
+  border-color: var(--accent);
+}