This repository hosts the Zama Bounty Program, an initiative designed to incentivize developers to contribute to the advancement of Fully Homomorphic Encryption (FHE) and Zama's open-source libraries. It targets developers interested in privacy-preserving computation, offering monetary rewards for innovative FHE applications and solutions.

How It Works

The program operates on a seasonal basis, with each season focusing on specific challenges related to Zama's FHE libraries like TFHE-rs, Concrete ML, and fhEVM. Submissions are evaluated based on code quality and, crucially, performance. The program emphasizes innovation and novel applications over simple bug fixes, aiming to accelerate FHE technology adoption.

Quick Start & Requirements

• Registration: Participants must register for specific bounties via a provided form.
• Submission: Code should be stored in a GitHub repository and submitted through a portal link received after registration.
• Resources: Bounties often leverage Zama's FHE libraries (TFHE-rs, Concrete ML, fhEVM). Benchmarks are run on Amazon EC2 HPC7A instances.
• Deadline: Season 8 submissions are due by May 18th, 2025.

Highlighted Details

• Season 8 bounties include developing a Fixed-point Arithmetic API using TFHE-rs ($10,000), an FHE-based Biological Age Estimation ML Model ($10,000), and a Confidential Benchmarking and Polling System on-chain using fhEVM ($10,000).
• The program has a public leaderboard showcasing top contributors and their earnings.
• Past winning solutions demonstrate applications like FHE-based AES-128, privacy-preserving image watermarking, and confidential auctions.

Maintenance & Community

• Support is available via email (bounty@zama.ai), Discord (discord.gg/zama), and a community forum (community.zama.ai).
• Specific bounty questions can be addressed by commenting on dedicated GitHub issues.

Licensing & Compatibility

• The README does not explicitly state the license for the bounty program itself or the submitted code. Zama's core libraries (TFHE-rs, Concrete ML, fhEVM) are typically open-source, but specific terms for bounty submissions and rewards should be reviewed in the program's terms and conditions.

Limitations & Caveats

• Submissions are reviewed at the end of each season, potentially leading to delayed feedback. Reward distribution is tiered, with the "best" submission receiving up to $5,000, and the program's terms and conditions should be consulted for full details on evaluation criteria and reward allocation.