The problem TrustLynk solves

Health insurance claim processing in India takes 15 to 90 days due to manual document verification, paper-based fraud, and siloed data between patients, hospitals, and insurers. ClaimChain eliminates this bottleneck by automating the entire pipeline - from verification to payout - using government health infrastructure, AI, and blockchain.

Challenges we ran into

1. ABDM Sandbox Instability
   Issue: The ABDM sandbox kept timing out and returning inconsistent response schemas during testing, breaking our verification flow mid-claim.
   Our Fix: Built a mock ABDM layer mirroring the real API structure, with retry logic and a fallback to manual review queue if the API fails.

2. Health Data on Blockchain — Legal Conflict
   Issue: We initially stored patient records directly on-chain. Midway through we realized this violates India's DPDP Act 2023 - blockchain data can't be deleted, conflicting with the right to erasure.
   Our Fix: Switched to a hash-and-pointer model - only the record hash, AI score, and verdict go on-chain. Actual health data stays encrypted off-chain.

3. Smart Contract Had No Circuit Breaker
   Issue: Our auto-payout contract had no way to stop if the reserve pool ran low — a mass-claim event could drain it completely with no intervention possible.
   Our Fix: Added a reserve threshold check, a daily payout cap of 20% of total reserves, and a multi-sig emergency pause requiring insurer + hospital + regulator to agree before freezing.