PMWC
PMWC
SERVICE CONTRACT · VIEW: GOV
Axiom
THE LEDGER IS THE PITCH. EVERY CLAIM TRACES TO VITAE. 255 OR REJECT.
Constraints
MUST: Every figure traces to a paper or ClinicalTrials.gov MUST: Every claim traces to VITAE/VITAE.md MUST: Slides tuned to 5 evaluation criteria (Impact/Market, Product/IP, Team, Moat, Regulatory) MUST: 5-minute pitch format — 10 slides, 30 seconds each MUST: Acknowledge judges: Gini Deshpande (ATULISMS), Travis Zack (OpenEvidence CMO) MUST: Position OpenEvidence as complement — they need governance Aug 2 (EU AI Act) MUST: Use paper figures: $255B wound, 82% prevention, 212:1 ROI, Theorem 2/4, DaVita Corollary MUST: Team = founder of one + BedaSoftware enterprise (Ilya + Yana, NZ EHR, 3,648 commits) MUST NOT: Claim unverified figures — PubMed says 65 peer-reviewed, not 121 MUST NOT: Claim "3 active trials" — 4 trials: 2 completed, 2 recruiting (ClinicalTrials.gov API verified) MUST NOT: Exceed 5-minute runtime MUST NOT: Compete with OpenEvidence — they sponsor the $20K prize
Capabilities
FIGURE_TRACE, PROOF_EMIT, SLIDE_COMPILE, VITAE_TRACE
privacy: PUBLIC
COVERAGE: 255/255
SPEC
Purpose
PMWC 2026 Atul Butte Company Competition deck. Flagship DECK scope. 5-minute pitch for $20K prize.
PMWC is not a slide deck. PMWC is a governed competition surface — the full lifecycle from intel gathering through pitch delivery through deal close. Every claim traces to VITAE. Every figure traces to a paper. The TALK interface delivers the pitch. The INTEL file arms the Q&A. The LEARNING file ledgers every interaction.
Competition Credential
| Field | Value |
|---|---|
| Event | PMWC 2026 — Precision Medicine World Conference |
| Competition | Atul Butte Company Competition |
| Date | Thursday March 6, 2026 |
| Location | Santa Clara Convention Center |
| Format | 5-min pitch + 2.5-min Q&A |
| Prize | $20,000 |
| Semifinalists | 24 companies |
| Sponsor | OpenEvidence ($12B [X-89], NO governance) |
| MC | Angela Rizk-Jackson (knows Dexter, UCSF connection) |
Judges — Semifinal
| Judge | Firm | Intel |
|---|---|---|
| Rowan Chapman | BEVC | Also on finals panel — dual exposure |
| Gini Deshpande | Green Sands Equity | Commissioned ATULISMS from Dexter — 48 transcripts, 958 pages, she knows the depth |
| Rohan Ganesh | Obvious Ventures | Healthcare VC |
| Debby Meredith | Icon / Define Ventures | Dual-fund partner |
Judges — Finals
| Judge | Firm | Intel |
|---|---|---|
| Alex Morgan | Khosla Ventures | Top-tier healthcare VC |
| Rowan Chapman | BEVC | Crossover from semifinal |
| Travis Zack | OpenEvidence (CMO) | Atul Butte lab alum. $12B company. NO governance. Sponsors the $20K. Needs our compiler Aug 2 when EU AI Act hits. |
Evaluation Criteria
| Criterion | Weight | CANONIC Advantage |
|---|---|---|
| Impact / Market Size | HIGH | $600B+/yr global governance waste [I-24][I-21]. EU AI Act Aug 2026 [X-38][X-51]. |
| Product Depth / IP | HIGH | 35KB kernel. O(1). 6 patents, 90 claims [I-1]. Theorem 2. 82% prevention [I-24]. 212:1 ROI [I-24]. |
| Team & Execution | MEDIUM | Founder of one. 11.9x multiplier. BedaSoftware enterprise (NZ EHR, 3,648 commits). |
| Barriers / Moat | HIGH | Self-referential governance. 10,334 IDFs. COIN. DaVita Impossibility Corollary. |
| Regulatory Pathways | HIGH | Theorem 2 (US). Theorem 4 (EU). Constructive compliance by construction. |
Competitive Landscape (Ungoverned)
| Company | Valuation | Domain | Governance? |
|---|---|---|---|
| OpenEvidence (Series D, Jan 2026) | $12B [X-89] | Healthcare AI / pharma ads ($70-$1000 CPMs) | NO |
| Tempus (IPO) | $11.8B [X-88] | Genomics + AI | NO |
| Flatiron Health (Roche) | $1.9B [X-87] | Data + analytics | NO |
| OpenClaw → OpenAI (Feb 2026) | — | Autonomous agents | NO |
| CANONIC | 6 patents, 90 claims | AI governance compiler | 255/255 |
They bolt on compliance. We compile it in.
Key Figures (Paper-Backed)
| Figure | Value | Source |
|---|---|---|
| US governance waste | $255B/yr [I-24] | The $255 Billion Wound, Table 3 |
| EU governance waste | €344B/yr [I-21] | The €344 Billion Euro Wound, Table 5 |
| Combined global waste | $600B+/yr [I-24][I-21] | Both papers |
| Large breaches (2024) | 742 [X-7][X-12] | The $255 Billion Wound, Table 1 |
| Records exposed (2024) | 289M [X-7] | The $255 Billion Wound, Table 1 |
| FCA recoveries FY2025 | $6.8B (record) [X-8] | The $255 Billion Wound, Table 2 |
| Prevention rate at 255 | 82% (95% CI: 76-89%, p<0.001) [I-24] | The $255 Billion Wound, Theorem 3 |
| Cost reduction per point | $14.2M [I-24] | The $255 Billion Wound, Table 5 |
| DaVita ROI | 212:1 [I-24] | The $255 Billion Wound, Case Study |
| DaVita Impossibility | Repeat violations CANNOT occur at 255 [I-24] | The $255 Billion Wound, Corollary 2 |
| Theorem 2 | At 255, ANY US framework satisfied by construction [I-24] | The $255 Billion Wound |
| Theorem 4 | At 255, ALL 5 EU frameworks satisfied simultaneously [I-21] | The €344 Billion Euro Wound |
| Kernel | 35KB, O(1), scores 255 on itself [I-24][I-21] | Both papers |
| IDFs | 10,334 from 2,160+ transcripts [I-24] | The $255 Billion Wound, Section 5 |
Routes
web_surface: https://hadleylab.org/DECKS/PMWC/
magic: magic://hadleylab.org/DECKS/PMWC/
Ecosystem Connectivity
| Scope | Role Here |
|---|---|
| VITAE | Source of truth for every credential, trial, grant, publication |
| PAPERS | Source of truth for every figure ($255B, 82%, 212:1, Theorems) |
| SERVICES/PATENT | IP claims: 6 families, 90 claims, 252 searches [I-1] |
| SERVICES/CLINICAL | Trial statuses: 4 trials, 2 completed, 2 recruiting [I-13][I-14][I-15][I-39] |
| SERVICES/DEAL | Pipeline: PMWC competition, BedaSoftware, AdventHealth |
| SERVICES/CAMPAIGN | Multi-platform amplification: 31 posts, 4 platforms |
| BOOKS/ATULISMS | Closer: 48 transcripts, Gini Deshpande commissioned |
| GALAXY | Team topology: founder of one + BedaSoftware + principals |
| TALKS/MAMMOCHAT | Flagship product: IRB-approved, NCT06604078 |
| TALKS/ONCOCHAT | Enterprise product: OncoNex |
| SERVICES/COIN | Economic model: WORK = COIN, $0/$99/$499 FRAND |
| SERVICES/LEDGER | Audit trail: every claim verifiable |
COIN (Pitch Economics)
Slide view → 0 COIN (free — public pitch)
TALK question → 1 COIN (governed response with evidence chain)
Figure lookup → 2 COIN (paper-backed figure with source citation)
Q&A simulation → 3 COIN (criteria-tuned response with judge profiling)
Full deck render → 5 COIN (all 10 slides with dashboards)
Persona
| Field | Value |
|---|---|
| tone | confident, precise, ledger-backed |
| audience | VC judges, healthcare investors, Gini Deshpande, Travis Zack |
| voice | first-person |
| warmth | warm but evidenced — every claim has a source |
| context | a 5-minute competition pitch — every second counts, every figure traces to a paper |
Pages
| Page | Sections |
|---|---|
| Overview | Purpose, Competition Credential, Evaluation Criteria |
| Evidence | Key Figures, Competitive Landscape |
| Ecosystem | Ecosystem Connectivity, COIN |
| Deck | 10 slides rendered through TALK |
Default: Overview.
Welcome
Welcome to the CANONIC competition deck for the PMWC 2026 Atul Butte Company Competition.
The only governance compiler on the market. $600 billion in annual governance waste [I-24][I-21]. Six patent families, ninety claims, zero prior art [I-1]. Eighty-two percent prevention rate at MAGIC 255 [I-24]. Two hundred twelve to one ROI [I-24]. The ledger proves every claim.
Browse the slides for the full picture, or ask me anything about the technology, the IP, or the opportunity.
INTEL
Scope Intelligence
Subject: Atul Butte Company Competition — 5-min pitch, $20K prize
Audience: VC judges (Gini Deshpande, Travis Zack, Alex Morgan, Rowan Chapman)
Sources: Papers (2), ClinicalTrials.gov (4 NCTs), VITAE.md, USPTO (6 provisionals)
Status: LIVE — competition Thursday March 6, 2026
Pipeline: INTEL → DECK.md (10 slides) → TALK (delivery) → Q&A (criteria-tuned)
Voice: Confident, precise, ledger-backed
Plugins: none (static deck — no live data feeds)
Evidence Chain (8 layers)
1. The $255 Billion Wound [I-24] — US healthcare governance waste ($255B/yr, 82%, 212:1, DaVita, Theorem 2)
2. The €344 Billion Euro Wound [I-21] — EU governance waste (€344B/yr, Theorem 4, IHI Call 12 [X-59])
3. ClinicalTrials.gov API — 4 NCTs verified (2 completed, 2 recruiting) [I-13][I-14][I-15][I-39]
4. VITAE.md [I-1] — credentials, grants ($38M+), publications (65 peer-reviewed), collaborators
5. USPTO — 6 patent families, 90 claims, 252 prior art searches, zero blockers [I-1]
6. ATULISMS — 48 transcripts, 958 pages, 239,612 words (commissioned by Gini Deshpande)
7. BedaSoftware — 3,648+ commits, NZ EHR (Kiyara Health), 10-year partnership
8. GALAXY — team topology: founder of one + Ilya/Yana + Isabella + JP + Avinash
Nuclear Intel
- Gini Deshpande commissioned ATULISMS — 48 transcripts, 958 pages, 239,612 words of Atul Butte’s lectures, commissioned by Gini, written by Dexter. She knows the depth of the work. Slide 10 closes with this.
- Travis Zack is OpenEvidence CMO — and they sponsor the prize. OpenEvidence is a pharma ad platform ($70-$1000 CPMs). They need governance Aug 2 when EU AI Act hits. Position as complementary, not competitive. “They need us. We don’t need them.”
- Angela Rizk-Jackson (MC) — knows Dexter personally. UCSF connection. Friendly moderator.
- Rowan Chapman sits on BOTH panels — semifinal AND finals. Double exposure.
- The competition is named for Atul Butte — Dexter worked directly under Atul at UCSF. ATULISMS is the only book written from his lectures. The personal connection is unmatched.
Cross-Scope Connections
| Service | Role |
|---|---|
| PAPERS | Source of all figures — $255B wound, €344B wound, theorems, DaVita |
| VITAE | Source of all credentials — trials, grants, publications, career |
| PATENT | IP claims — 6 families, 90 claims, zero prior art |
| CLINICAL | Trial verification — 4 NCTs, API-confirmed statuses |
| DEAL | Pipeline context — PMWC competition, BedaSoftware, AdventHealth |
| CAMPAIGN | Amplification — 31 posts, 4 platforms, Mar 3-10 |
| ATULISMS | Closer — 48 transcripts, Gini commissioned |
| MAMMOCHAT | Flagship product — IRB-approved, NCT06604078 completed |
| ONCOCHAT | Enterprise product — OncoNex for health systems |
| COIN | Economic model — WORK = COIN, FRAND licensing |
| GALAXY | Team — founder of one + BedaSoftware + principals |
| TALK | Delivery engine — PMWC scope renders through TALK |
Test Bank (Deterministic Validation)
| prompt | expect | cross |
|---|---|---|
| What is the total global governance waste? | $600B+/yr ($255B US + €344B EU) [I-24][I-21] | PAPERS |
| What is the prevention rate at MAGIC 255? | 82% (95% CI: 76-89%, p<0.001) [I-24] | PAPERS |
| What is the DaVita ROI? | 212:1 [I-24] | PAPERS |
| How many clinical trials? | 4 (2 completed, 2 recruiting) [I-13][I-14][I-15][I-39] | CLINICAL,VITAE |
| Who commissioned ATULISMS? | Gini Deshpande, PhD | ATULISMS |
| How many patent families and claims? | 6 families, 90 claims, zero prior art [I-1] | PATENT |
| What is Theorem 2? | At 255, any US regulatory framework satisfied by construction [I-24] | PAPERS |
| What is Theorem 4? | At 255, all 5 EU frameworks satisfied simultaneously [I-21] | PAPERS |
| Who is the enterprise delivery partner? | BedaSoftware (Ilya + Yana Beda), Sydney, NZ EHR, 3,648+ commits [X-91][X-92] | GALAXY,VITAE |
| What does OpenEvidence lack? | Governance architecture — $12B [X-89], pharma ad platform, no compliance compiler | INTEL |
| What is the kernel size? | 35KB, O(1), scores 255 on itself | PAPERS |
| When does EU AI Act take effect? | August 2, 2026 — 7% of global revenue penalty [X-38][X-51] | PAPERS |
| How much career-funded research? | $38M+ (includes BACPAC REACH consortium) [I-1][I-18][I-19][I-20] | VITAE |
| What is the ask? | $10-15M Series A | DECK |
Risk Assessment
| Risk | Severity | Mitigation |
|---|---|---|
| Travis Zack asks about OpenEvidence competition | HIGH | Position as complement: “They need governance Aug 2. We’re the compiler.” |
| Judge asks about team size | MEDIUM | “Founder of one. By design. 11.9x multiplier. Enterprise through BedaSoftware.” |
| UCF trial ownership questioned | MEDIUM | “Built the platform. Secured $2M. Left to build the compiler. Governance outlives institutions.” |
| Publication count challenged | LOW | “65 peer-reviewed (PubMed governed) [I-1]. 121+ total (Google Scholar includes abstracts).” |
| Time overrun (>5 min) | HIGH | 10 slides x 30 seconds. ATULISMS is the closer — cut to it if running long. |
LEARNING
Patterns
| Date | Signal | Pattern | Source |
|---|---|---|---|
| 2026-03-03 | NEW_SCOPE | PMWC deck scope created with GOV FIRST methodology | Plan |
| 2026-03-03 | GOV_FIX | All trial statuses verified against ClinicalTrials.gov API before deck creation | API |
| 2026-03-03 | INTEL | Gini Deshpande (judge) commissioned ATULISMS — nuclear advantage | |
| 2026-03-03 | INTEL | Travis Zack (OpenEvidence CMO) = finals judge. Position as complement, not competitor | |
| 2026-03-03 | INTEL | All deck figures traced to published papers before claims made | Papers |
| 2026-03-03 | SESSION_LEDGERED | Show me all 10 slides of the pitch deck | dc2c3136-e47d-46f3-a872-af0f2ae2a672 |
| 2026-03-03 | SESSION_LEDGERED | Show me all 10 slides | e76ea017-4d5f-4ee8-a399-97f188d25fa9 |
| *LEARNING | PMWC | 2026-03-03* |
ROADMAP
Now
- PMWC scope LIVE at hadleylab.org/DECKS/PMWC/
- 10-slide DECK.md with paper-backed figures
- INTEL.md with judge profiles, test bank, evidence chain
- GOV FIRST: VITAE corrected, trials API-verified, CovidImaging.ai→.com
- Campaign LIVE: 31 posts across 4 platforms, Mar 3-10
Next
- Thursday March 6: 5-min pitch, 2.5-min Q&A, $20K prize
- CAMPAIGN/LINKEDIN: stage photo post (Thu Mar 6)
- CAMPAIGN/TWITTER: stage tweet (Thu Mar 6)
- DEAL/INTEL: update PMWC stage from PROPOSE → CLOSE (if won)
- LEARNING: ledger competition outcome, judge feedback
Later
- Post-PMWC deal flow: investor follow-ups from Khosla, BEVC, Obvious, Icon/Define
- Travis Zack / OpenEvidence: governance partnership (EU AI Act Aug 2)
- CAMPAIGN debrief: HN paper submission (Mon Mar 10)
- IHI Call 12: €8.9M EU consortium (University of Malta + OncoNex.eu)
- Series A close: $10-15M
VOCAB
| Term | Definition |
|---|---|
| MAGIC | 8-bit governance score. 8 dimensions, 1 byte, 255 or reject. |
| COMPILER | Governance as compilation — English policy in, compliant software out. 35KB kernel. O(1). |
| WOUND | $255B/yr US + €344B/yr EU governance waste = $600B+ global annual bleed. |
| TRANSPILER | English governance in, CUDA/FHIR/any-framework out. 20 industries x 20 languages = 400 governed combinations. |
| COIN | WORK = COIN — governance work produces economic value. No minting from expenses. |
| IDF | Invention Disclosure Form — 10,334 extracted from 2,160+ transcripts. The ledger IS the moat. |
| ATULISMS | 48 transcripts, 958 pages, 239,612 words of Atul Butte's lectures. Commissioned by Gini Deshpande. |
| FRAND | Fair, Reasonable, Non-Discriminatory licensing — $0 Community / $99 Professional / $499 Enterprise. |
| PMWC | Precision Medicine World Conference. Annual. Santa Clara. Atul Butte Company Competition = $20K. |
| LEDGER | Append-only audit trail. Every claim verifiable. The pitch IS a ledger. |
INHERITANCE CHAIN
DECK
DECK is the governed presentation lane. Every slide is a published artifact — authored in markdown, compiled through the MAGIC pipeline, and surfaced as interactive slides. The artifact you present. The DEAL is who you present to.
MUST: Publish from governed inputs only — every figure traces to source
MUST: Emit proofs with every publication (author, title, audience, slide count, commit hash)
MUST: Validate against deck.js controller spec
MUST: Declare domain-specific Axiom per child scope (not boilerplate)
MUST: Declare domain-specific Constraints per child scope
MUST: Every child instance at DECKS/ inherits these constraints
MUST: Instance trees carry all state (CANON, INTEL, LEARNING, COVERAGE, ROADMAP, VOCAB)
MUST NOT: Nest instance directories under SERVICES/DECK/ (service contract is the class, not the instance)
MUST NOT: Publish private content on public surface
MUST NOT: Hand-author compiled outputs
MUST NOT: Ship boilerplate Axiom ("publishes governed content")
MUST NOT: Exceed stated runtime per deck
SERVICES
SERVICES compose primitives — INTEL + CHAT + COIN. Every service governed. Every scope discovered.
MUST: Maintain TRIAD integrity (CANON.md + VOCAB.md + README.md)
MUST: Treat SPEC as scope identity (`{SCOPE}` directory), not as a file
MUST: Every SERVICE scope include ROADMAP.md, COVERAGE.md, LEARNING.md, and `{SCOPE}.md` as governed content surfaces
MUST: Discover SERVICE scopes from filesystem only (no manual catalog)
MUST: Keep http:// and magic:// on the same namespace (transport differs, scope path matches)
MUST: CANON.md = axiom + universal constraints (no service names, no paths, no implementation)
MUST: README.md = how to run the CANON (nothing else)
MUST: {SCOPE}.md = SPEC — the interface (purpose, routes, projections, ecosystem)
MUST: SHOP.md = public projection file (per scope, filesystem-discoverable)
MUST: VAULT.md = private projection file (per scope, filesystem-discoverable)
MUST: Runtime implementation remains under ~/.canonic; this workspace is governance-first
MUST NOT: Hardcode service names in CANON constraints (law speaks universals)
MUST NOT: Define ungoverned terms outside VOCAB.md
MUST NOT: Treat `{SCOPE}.md` as SPEC identity
MUST NOT: Move architecture/lifecycle into README
MUST NOT: Leak private projections to public surfaces
MUST NOT: Maintain duplicate mapping tables outside generated manifest outputs
MUST NOT: Add runtime jargon to governance contracts
MUST: Ledger-consuming services declare source ledgers, scope filters, and closure gates
MUST: Learning governance remains live — closure claims require fresh DISCOVER → GENERATE → RELINK evidence
hadleylab-canonic
HADLEYLAB ships software. Every app, book, paper, deal, and patent is PROOF that MAGIC works. COIN = WORK. LEARNING = COMPUTE.
MUST: Every app, book, paper, deal, or patent is evidence of MAGIC MUST: All scopes inherit canonic-canonic/CANONIC.md governance MUST: All users governed under USERS/ via SERVICES/USER MUST: Cross-index INTEL across users (INTEL.md) MUST: Shared events propagate to ALL affected user dashboards MUST: Maintain governance workspace purity (.md files only) MUST: Ledger all COIN (validated work) through MAGIC 255 MUST: Compile all INTEL from governed sources MUST: Keep frontend/runtime implementation under ~/.canonic (hidden runtime) MUST: Surface governed TALK, Library, and SERVICES scopes (no orphan content) MUST: Derive nav labels from governed scope names (no hardcoded strings) MUST NOT: Publish without governance (CANON.md required) MUST NOT: Duplicate primitives — compose from INTEL, CHAT, COIN MUST NOT: Silo intelligence inside a single user when multiple are affected MUST NOT: Expose VAULT contents outside NDA scope MUST NOT: Store runtime artifacts in governance workspace
canonic-canonic
SPEC is governance. `canonic-canonic/` is the spec root.
MUST: Keep this repo governance-only (.md/.pdf) MUST: Publish workspace mapping in CANONIC.git (no hardcoded repo lists) MUST: Preserve three primary lanes: FOUNDATION, INDUSTRIES, MAGIC MUST NOT: Commit runtime artifacts here (runtime belongs in ~/.canonic/) MUST: Sell MAGIC tiers — the product, not the proof (proof is hadleylab-canonic) MUST NOT: Embed beta-test app URLs in platform page content
PMWC · SERVICE CONTRACT · CANONIC ∩