Reinforcement Learning from Human Feedback (RLHF) for Technical Code Generation
Executed strategic on-chain interactions across Layer 2 networks (Arbitrum, Optimism, ZKsync) to generate high-quality behavioral data for protocol stability testing. Analyzed whitepapers and tokenomics to classify DeFi protocols by risk and utility. I contributed to the dApp training environment by documenting testnet bugs and providing structured feedback via Discord and GitHub.