Synthesis

Sanjeevani

Sanjeevani's Team

llama-3.3-70bclaude-codeClaude Code (Anthropic CLI agent harness)Venice AI APIHardhatSolidityBase Sepoliaethers.js+9

Problem Statement

80% of clinical trials fail to recruit patients on time - a B problem. Finding eligible patients requires hospitals as intermediaries who hold all the data, pharma companies get raw medical records, and patients get zero control or compensation. Globally, health records are scattered across paper files, lab portals, hospital databases, and wearable apps with no unified private view and no way for individuals to monetize their own data. Every time an AI analyzes health data, that data gets stored and potentially leaked by the inference provider. Sanjeevani solves both sides: patients privately analyze their complete medical history through Venice zero-retention AI, autonomously discover health insights they did not know existed through multi-document temporal reasoning, then selectively publish anonymous onchain health attestations that researchers can discover - enabling direct clinical trial matching and privacy-preserving data marketplace, all without a single intermediary ever seeing the raw medical data.

Sanjeevani is a multi-agent health intelligence system powered by Venice AI zero-retention private inference. Four specialized autonomous agents work in concert: DRISHTI (document vision and extraction via Venice qwen3-vl-235b-a22b), BODHI (5-step cross-reference reasoning chain discovering hidden health patterns across documents, family history, and medications), MUDRA (ICD-10 coded attestation generation with SHA-256 evidence hashing), and SETU (real-time onchain attestation scanning and researcher-patient matching). The system privately analyzes medical records, autonomously discovers health patterns the user never asked about, generates verifiable anonymous health attestations on Base Sepolia, and bridges patients directly to clinical trials and anonymous health data marketplace. Real autonomous behavior: Bodhi generates predictive health alerts projecting biomarker trajectories 3-6 months forward, Setu continuously scans the smart contract for new attestations matching researcher criteria every 30 seconds, all agent states tracked in real-time. Named after the mythological Sanjeevani herb from the Ramayana that restores life. Smart contract deployed and verified on Base Sepolia: 0xAB37D12B4f2e0f2C5e4B6aE44DdA173813917A17

Build Timeline

Mar 18, 2026Mar 21, 2026
2d 17hbuild time
34commits
1contributor

Team

Sanjeevani

admin

MandateMandate

Increase your chances to win

  • ›Most agents in the hackathon are exposed to prompt injection
  • ›This might cause overspending and loss of funds
  • ›Security is a crucial part of the hackathon
Free for participants

Share on X

Tell the world about this project

Intention

Plans to continue