Background in test automation, platform engineering, and developer tooling. Currently going deep on LLM apps, RAG, agentic systems, and evaluation — the bits that separate working notebooks from production AI.
I've spent my career making engineering teams ship faster — designing QA platforms, automation frameworks, and developer tooling. I tend to build the systems that run the tests, not just the tests themselves.
That same systems-thinking now points at AI. The hard problems I care about aren't "can this LLM answer my question" but "how do you build something that answers reliably at scale" — chunking strategy, retrieval quality, eval pipelines, cost control, failure mode analysis. SDET-brain meets AI engineering.
I write every line of code myself, deploy to my own infra, and treat evaluation as a first-class concern. The QA background is a real edge in this space, not a previous chapter.
Self-hosted personal knowledge base. Built end-to-end with Next.js and Express. You're looking at it.
A 6-phase, 16-week roadmap covering LLM core, RAG, agents, evaluation, production deployment, and AI system design.
Years of building test frameworks, automation pipelines, and developer tooling. The systems that ran the tests, not just the tests themselves.
Hiring for AI engineering, applied AI, dev tools, or platform roles? Or want to compare notes on RAG / eval / agentic systems? I read every email.