Code Metal

Own one or more AI platform components from design through production operation.Build and operate model-serving, agentic infrastructure, context-engineering, observability, or experimentation services.Write clean, modular, well-tested production code and review teammates’ code and designs.Debug production issues such as tail latency, GPU memory pressure, and failing or looping agent

Set the technical direction and architecture for the AI platform, lead the four-person team, own design documents and RFCs, guide build-versus-buy decisions, and mentor engineers.Deploy, benchmark, and tune production inference for open-weight models on vLLM, SGLang, and TensorRT-LLM.Own the model gateway for self-hosted and commercial models, including authentication, routing, fail

Assess validation practices, identify gaps, and establish scalable quality strategies for code-generation, transpilation, and software-transformation workflows.Build automated quality and regression infrastructure for AI-assisted software workflows that produces reproducible engineering evidence.Standardize quality signals and release evidence, including regression trends, failure a