All projects
Insurance · Agentic AI

Recoupe — Autonomous Subrogation

Seven agents that read a closed claim, assign fault by jurisdiction, compute what is recoverable, and pursue it — every decision citation-grounded and auditable.

The problem

US property & casualty insurers leave an estimated $15–25B in subrogation recovery on the table every year — not because they cannot recover it, but because human adjusters can only work the biggest files. The long tail of small claims gets dropped at intake.

Subrogation is an ideal agentic testbed: the ground truth is codifiable (negligence law is published, carrier behaviour is observable, recoverable amounts are derivable), and the decisions repeat with the same shape every time.

Architecture

  1. 1

    Intake

    Reads the claim file and extracts parties, losses, and fault facts (LLM extraction or deterministic heuristics).

  2. 2

    Liability

    Assigns the fault percentage under the correct state's negligence regime (comparative / modified / contributory).

  3. 3

    Quantum

    Computes the recoverable dollar amount given fault, damages, and policy limits.

  4. 4

    Strategy

    Decides pursue or drop, with the threshold tunable per carrier.

  5. 5

    Demand

    Drafts the demand letter with grounded statutory citations.

  6. 6

    Negotiation

    Works counter-offers against carrier-specific settlement behaviour.

  7. 7

    Litigation

    Escalates only when the expected value of suit beats settlement.

  8. 8

    Audit trail

    Every decision appended with model, confidence, and evidence; streamed live to the UI over SSE.

Key tradeoffs

Deterministic skeleton, LLM polish — the math is codified Python; the model extracts and narrates.

Why · Insurance is regulated. A system that produces different fault percentages run to run is not deployable; output is bit-identical without a key.

A citation-integrity guardrail rejects unsourced authorities before they reach the audit trail.

Why · Lawyers do not hire researchers who cite cases that do not exist; AI generating legal arguments should meet the same bar.

The codified knowledge base (per-jurisdiction negligence rules + carrier graph) is the moat, not the agent chain.

Why · Anyone can wire seven LLM calls together; almost nobody builds the per-jurisdiction map underneath.

Audit trail as a first-class product feature.

Why · The trail — model, confidence, evidence, approver — is what makes a compliance officer say yes.

Eval results

measured
Recovery rate

Actual recovered ÷ truly recoverable, on synthetic claims with known-true values.

measured (pts)
Liability MAE

Mean absolute error in the fault percentage vs known truth.

measured %
Quantum error

Mean error on the recoverable dollar amount.

tracked
Citation integrity

Share of cited authorities that were genuinely retrieved.

Production proof

The artifact that keeps the numbers honest — the eval harness / monitoring gates that run in CI, not a one-off notebook result.

Self-grading on ground truth

CI · passing
Liability MAE scored
Quantum error scored
Citation integrity scored

Synthetic claims carry known-true fault and recoverable values, so every agent decision is scored — most agentic demos have no quantitative answer to "how right is it?"

Targets a real $15–25B annual leak with an auditable, citation-grounded pipeline — the long-tail claims humans cannot afford to chase, pursued at machine cost.

Let's talk

I'm focused on finance AI — credit risk, RegTech, AML, and agentic investment research. Open to roles, mentorship, and collaborators in fintech, quant, and bank AI.