Richard Atkins

Python, TypeScript/Node, SQL. Building LLM apps: agents, RAG, evals, guardrails, OTel tracing. Hosted + local (Ollama) inference. Generalist: 20+ yrs across backend, front end, cloud.

The economics of running AI in production: cost, latency, and where local inference beats cloud. Platform patterns for shipping AI safely at scale. And how AI reshapes org design and leadership.

Open-source AI reference repos: agent-eval-starter (evals + observability as a CI gate) and resumable-llm-pipeline (resumable, cost-aware LLM pipelines). Plus a 5-part series on AI and leadership.

AI and platform transformation, architecture reviews, and scaling engineering teams. Mentoring engineers. Speaking or writing on the human side of AI at work. Reach out if that's you.