truptis

Designed SSV-based distributed validator technology for 200+ validators, implementing operator management, cluster coordination, and slashing protection mechanisms • Developed staking and restaking systems integrating EigenLayer architecture with delegation, withdrawal credentials, and reward accrual logic • Engineered Ethereum Pectra-compliant 0x02 withdrawal credentials with custom validation logic, reducing withdrawal processing time by 40% with automated reward distribution. Led design and implementation of modular multi-agent architecture using AutoGen and LLMs for DEX, enabling autonomous decision-making and improving user profitability by 30%+. Integrated on-chain data (subgraphs, liquidity pools) with off-chain sources (CoinGecko, market APIs) using RAG techniques to reduce AI hallucinations. Coordinated with frontend, backend, and product teams to deliver seamless AI integration via FastAPI endpoints. Mentored 2 junior developers and conducted weekly knowledge-sharing sessions on AI/blockchain integration best practices.

Engineered LayerZero-based cross-chain governance system with custom storage proofs, increasing multi-chain participation by 62%. Led integration of 6+ blockchain networks (Arbitrum, zkSync Era, Optimism, Ethereum, BSC, Kava) for seamless DeFi asset transfers. Designed cross-chain oracle framework ensuring consistent risk parameters across all chains, reducing protocol-wide risk. Created comprehensive technical documentation and trained 15+ team members on cross-chain development patterns.

Conducted smart contract security audits identifying 15+ critical vulnerabilities, reducing protocol risk. Developed automated risk monitoring systems generating governance proposals, reducing manual oversight time. Designed secure proposal validation framework with automated simulations, achieving 100% success rate for production deployments. Collaborated with product and business development teams to align technical capabilities with market requirements.

Built real-time emotion detection system using CNNs achieving accuracy on video streams. Developed 'Attend-Ease' face recognition system improving attendance tracking efficiency.