Tom — AI Engineering Leader & Independent Consultant 13 years building AI and ML systems at scale — Amazon, Lyft, Turing — across infrastructure, developer tooling, and applied ML. Currently pursuing a DBA researching agentic memory systems (multi-agent RAG architectures, topology-aware retrieval). I help engineering teams and founders with three things: * AI architecture that ships: Not proof-of-concept demos — production systems with real retrieval pipelines, agent orchestration, and evaluation frameworks. I've designed and built systems spanning code intelligence (AST-aware retrieval with dependency graphs), autonomous optimization engines, and voice identity platforms. * Technical leadership without the overhead: Fractional director-level guidance for teams scaling from prototype to production. Hiring strategy, CoE design, build-vs-buy decisions, vendor evaluation. I've led 150+ member engineering orgs and know what breaks at each growth stage. * First-principles AI education: I maintain no-magic — 48 single-file, zero-dependency ML implementations with 1,200+ GitHub stars. If you're a founder or PM who needs to understand what's actually happening inside the models your team is building on, I can get you there without the hand-waving. * Open-source work: archex (codebase intelligence), armory (1-6 production skill packages for AI coding agents), anneal (autonomous optimization), VaultMind (knowledge management). I don't do hype. I build things, break them, and know the difference between what works in a demo and what survives prod traffic. Book a call if you want straight answers, not a pitch deck.