I build agent systems people can actually rely on.

For the past few years I’ve worked across the stack of applied AI — starting in software engineering, then moving into AI engineering, where I now build agentic platforms, no-code agent builders, voice agents and the multi-modal data pipelines that feed them.

The thread through all of it is reliability. Anyone can get a model to do something impressive once. I care about the harder, less glamorous part — the evals, guardrails, tracing and architecture that decide whether an agent keeps working when real users and real data hit it. That’s the difference between a demo and a system you can sell.