Cambrian
Every year, over a million people die on the roads and fifty million more are injured.
Core trick
Extracts structured facts from injury records, cross-checks dates and amounts across sources, and re-derives every petition figure in a separate verifier pass.
Description
Every year, over a million people die on the roads and fifty million more are injured. In India alone, over a million compensation cases are stuck in tribunals — ten billion dollars unpaid — with ninety percent stalled by documentation problems: missing facts, wrong arithmetic, unreconciled records.
Cambrian is a living knowledge base for catastrophic injury victims. It ingests police reports, hospital records, and financial documents, extracts structured facts, and reconciles them across sources — catching conflicts like mismatched dates between a police FIR and a hospital record. Contradictions are flagged, never buried. The legal agent is called Nyayasetu.
Two independent verification agents — each in a fresh context with no memory of prior stages — audit the output. A KB verifier grades against 10 invariants. A petition verifier re-derives every figure from scratch. If any number is wrong, it rejects and the drafter revises autonomously.
The legal petition is one spoke. The same KB powers medical advocacy, rehab planning, and AAC communication for patients who've lost the ability to speak.
README excerpt
# Cambrian
A **living knowledge base for catastrophic injury** — powering legal drafting,
medical advocacy, rehab planning, and AAC communication from a single, self-maintaining
case record.
> Every 24 seconds, someone dies on the world's roads. Over 50 million more are
> injured each year. In India alone: 155,000+ annual deaths, over a million MACT
> cases stuck in tribunals, $10 billion in unpaid compensation, and 90% of claims
> stalled by documentation problems. Families face this while already dealing with
> the trauma of losing — or watching — a loved one suffer.
Cambrian ingests every document in a catastrophic injury case — police reports,
hospital records, disability certificates, salary slips, insurance papers — and
builds a **structured, sourced, self-consistent knowledge base**. From that one KB,
multiple agents act:
- **Legal drafting** — MACT petition with every rupee traced to a sourced fact
- **Medical advocacy** — treatment timeline, prognosis tracking, second-opinion prep
- **Rehab planning** — milestone tracking, therapy scheduling, equipment needs
- **AAC communication** — for patients who've lost the ability to speak
Today's demo shows **NyayaSetu**, the legal agent. The architecture generalizes to every spoke.
## Architecture
```
Documents (9+)
│
▼
┌──────────┐ ┌─────────────┐ ┌──────────────┐ ┌───────────┐ ┌─────────────────┐
│ Ingest │───▶│ Reconcile │───▶│ Verify KB │───▶│ Draft │───▶│ Verify Petition │
│