I build production-ready agentic AI systems, multi-agent architectures, and RAG pipelines, plus the interfaces that make them usable. This site is where I write it down: the architecture decisions, the patterns that worked, and the mistakes worth avoiding.
The hard part of AI engineering is rarely the model. It's the system around it: retrieval, evaluation, and the unglamorous reliability work that makes it trustworthy.
Building
In production