I'm Josh Zirena, a tech lead focused on production AI systems. I've worked on RAG assistants, LLM observability, guardrails, backend APIs, AWS infrastructure, and AI platform integrations.

Most AI work fails the same way: an impressive demo that quietly falls apart when it meets real users, real data, and real reliability requirements. I specialize in that gap: the architecture, evaluation, guardrails, and observability that turn a prototype into something a team can actually run.

I care about systems that are grounded (they don't make things up), measurable (you know when they get better or worse), and maintainable (your team can extend them after I'm gone). I try to leave every team more capable than I found it.