I build AI products that work in production — not just in demos. Currently engineering AI systems at Provility. 10+ years across government, healthcare, fintech, and public transit means I know what it takes to ship systems that hold up under real conditions.
10+ years of production engineering, applied to the full stack of AI product development.
Designing production RAG pipelines that reduce hallucinations and improve answer quality — from embedding strategies and chunking to retrieval optimization, re-ranking, and grounding.
Building agent architectures, tool-calling systems, and multi-step reasoning workflows that connect LLMs to real business operations — APIs, retrieval pipelines, and autonomous task execution.
Engineering low-latency LLM streaming systems via SSE, optimising token throughput and caching, and building real-time AI experiences that perform under production load — not just benchmarks.
End-to-end delivery of AI-powered products — scalable backend systems, well-designed APIs, and the frontend experiences that make AI genuinely usable. From architecture decisions to production deployment.
