DeFi · Open Source
Built a cyclic DEX arbitrage detector in Rust that forks live Ethereum mainnet via REVM for bit-exact swap simulation. Features an actor-based shared RPC backend for thread-safe concurrent simulations, ternary search optimization for optimal trade sizing, and a generic SwapSimulator trait for pluggable execution backends.
REVM + Alloy + Rayon: production-grade EVM simulation in Rust
Rust, REVM, Tokio, MEV
NFT
On-chain NFT auctions are prohibitively expensive. Built a hybrid architecture using ECDSA signatures for off-chain bidding with on-chain settlement, achieving trustless ERC-721 ↔ ERC-20 swaps at a fraction of the gas cost.
90% gas savings vs. traditional on-chain auction patterns
Solidity, TypeScript, Foundry, ECDSA
DeFi
Managing concentrated liquidity positions on V3 AMMs requires precise tick math and multi-step transactions. Built a production-grade TypeScript toolkit for minting, collecting fees, and rebalancing SushiSwap V3 positions programmatically.
Full V3 lifecycle management: mint, collect, rebalance
TypeScript, Ethers.js, Uniswap V3, DeFi
SDK · Open Source
GenLayer needed a developer-friendly SDK to onboard builders to its AI-powered blockchain. Designed and shipped the primary Python SDK with web3.py integration, strong typing, and comprehensive documentation.
Primary SDK for GenLayer developer ecosystem
Python, Web3, AI, SDK
SDK · Open Source
Intelligent contracts with AI execution need deterministic testing infrastructure. Built a Pytest-based testing suite that enables CI/CD pipelines for GenLayer's AI-powered smart contracts with reproducible results.
Enables CI/CD for intelligent contracts
Python, Pytest, Testing, AI
Enterprise
Enterprise HR platform struggling with API latency and growing frontend bundle. Led performance optimization across backend (query refactoring, Redis caching) and frontend (Vue.js migration), delivering measurable improvements.
35% API performance improvement for 5,000+ org platform
React, Vue.js, Ruby on Rails, AWS
AgTech
Agronomists across Latin America needed reliable field data collection and real-time crop analytics. Built backend APIs and geospatial data pipelines supporting mobile and web clients with high availability.
Serving 1,500+ agronomists across 3 countries
Django, Python, PostgreSQL, AWS