I build AI systems that actually work in production, which is harder than most people think.
Right now I'm at ExperioLabs designing a multi-phase AI agent on LangGraph — the kind where making it work is week one, and making it work every time is the next three months, because the hard problems aren't the AI, they're everything else. Before that I built a newsletter platform that handles millions of subscribers across 2,000+ newsletters, and a handful of other SaaS products from zero to shipped.
My career arc is competitive programmer → full-stack engineer → AI engineer. The competitive programming background turns out to be surprisingly useful when you're designing graph-enforced agent workflows — it's the same kind of problem decomposition, just with more YAML.
I've been writing production code for 8+ years across Python, TypeScript, React, Node.js, Django, LangChain, Neo4j, and whatever else the problem demands.
