AI architect. Building reliable intelligent systems.
I care about AI that's reliable, not just impressive. AI that does the unglamorous work so people can focus on higher goals and creative pursuits. The moat in AI isn't the model. It's architecture, evals, and how deeply you understand the domain you're building for. I'm drawn to the parts most people skip i.e. the infrastructure and evaluation layers that decide whether an AI system actually holds up when real users depend on it.
What I'm currently focused on.
A 6-week on-site residency in Bangalore for going deep on AI research. I'm exploring negotiation strategies and Game theory behaviour in agent to agent interactions. This is pathway to building Agent-2-human negotiation and evaluating which strategies work better.
Building vertical AI solving messy workflows in ops layer, starting with supply chain. 135 million driver hours lost to detention annually, financial impact $15B. I'm working on getting those hours back for trucking companies