Summary
John Munger is an AI platform engineer with 11 years of experience who designs control-plane architectures, agent workflows, and retrieval-augmented systems to bring LLMs safely into production. He specializes in the layer around models—orchestration, policy, and execution—building TypeScript runtimes that enforce bounded execution, validation gates, and multi-model routing for cost and latency tradeoffs. His platform background in release control, versioning, and failure prevention informs pragmatic designs like RAG pipelines with pgvector, CI/CD agent frameworks with rollback safeguards, and deterministically constrained tool use. Based in Denver, he favors safe failure over autonomous behavior and focuses on observability and predictable behavior in real applications, not demos. An underrated strength is translating enterprise-grade release and dependency controls into AI runtimes that minimize downstream production incidents.
11 years of coding experience
8 years of employment as a software developer
BA, Web and Interactive Development, BA, Web and Interactive Development at Art Institute of Colorado
Bachelor of Applied Science - BASc, Web Application and Interface Design, Bachelor of Applied Science - BASc, Web Application and Interface Design at Texas State Technical College