Low-latency trading infrastructure in Rust
Market data · mark-pricing oracles · on-chain execution across CEX and DEX venues
Most of my work lives in private repositories. This profile is a pointer, not a portfolio.
📡 Low-latency market data Solo-built a ~250k-LOC market-data and pricing-oracle platform in Rust: 45 crypto venue-feeds across 7 chains normalized into one event model, feeding an Avellaneda–Stoikov mark-pricing engine with triple-transport egress (shared-memory ring / UDP / NATS Core). Benchmarked ~20µs p50 producer wire-to-publish; ~7µs p50 delivery to co-located shared-memory consumers.
🦄 On-chain market making Production Rust LP market-making engine for concentrated-liquidity DEXs (Uniswap V3/V4, Aerodrome), live on Base, Arbitrum and Optimism through four custody backends including Fireblocks MPC and Zodiac Roles v2. Integer-exact concentrated-liquidity math implemented from scratch, atomic single-tx repositioning, oracle-gated adverse-selection control, and gas-EV bounded execution.
◎ Solana execution Near-zero-slot DEX execution via ShredStream/gRPC decoding and Jito/SWQoS transaction routing across 10+ Solana DEXs and launchpads, with MEV market-making and cross-pool arbitrage strategies.
🛡 Risk & execution quality Sub-5ms risk loop with automated circuit breakers across 15+ perpetual futures venues; post-fill markout attribution, effective and realized spread, VPIN, Kyle's λ and price-impact decomposition materialized in ClickHouse.
Core
Trading systems
Systems & latency
Data
Chain
DEX
Infrastructure
Dashboards — order books, execution quality, microstructure, venue intelligence
Background in aerospace engineering · MSc Innovation & Technology Management, University of Bath
