AI Platform Engineer

Code Metal is the leader in automated software engineering you can trust. As AI writes more of the world's code, the bottleneck in software has shifted from writing code to verifying it works, and AI cannot verify its own work with certainty. Code Metal takes a fundamentally different approach: constrain AI to what it does reliably, verify every step independently of the model using formal methods, and keep engineers in the loop on the decisions that matter.

The result isn't code that probably works — it's code that is provably correct, with auditable proof. Customers including the U.S. Air Force, L3

Harris, RTX, and Toshiba use Code Metal to modernize legacy code, optimize performance on real hardware, and move prototypes to production, fast. Founded in 2023 with offices in Boston and San Francisco, Code Metal isfunded by Accel, Salesforce Ventures, B Capital, Smith Point Capital, J2 Ventures, Shield Capital, Overmatch, RTX, and others.
Learn more .

The Role

Code Metal's engineering teams are building AI-driven code transpilation and AI-enabled mission planning and wargaming. Both need the same foundations: models to serve, agents to run, context to manage, and results to measure. Our AI Platform team builds those foundations.

As a Staff AI Platform Engineer, you'll be the technical lead of this new four-person team. You'll architect and build the AI enablement stack our engineers depend on, from GPU inference serving and a model gateway up through agent harnesses, context engineering, observability, and AI experimentation management. It starts as an internal platform, but we're building it to product standard.

This is an engineering role first. Most of your time goes to designing, building, and operating production systems. You'll also need solid data science and AI research fundamentals: you'll work closely with our Applied AI Research team, and you'll sometimes run experiments yourself when a platform decision needs evidence.

Core Responsibilities

• Set the technical direction and architecture for Code Metal's AI platform and lead the team building it. Own the design docs and RFCs, help with build-vs-buy decisions, and mentor the team.

• Deploy, benchmark, and tune production inference for open-weight models on vLLM, SGLang, and TensorRT-LLM.

• Own the model gateway that teams use to reach self-hosted and commercial models, with consistent auth, routing, failover, quotas, and cost attribution.

• Design reusable agent harnesses and orchestration primitives that product teams can compose into reliable, verifiable workflows instead of rebuilding them for each product.

• Build context-engineering services for memory, retrieval, and data discovery, so agents get the right information within their context and cost budgets.

• Instrument the stack end to end with Open Telemetry traces and service metrics, and build the experiment-tracking and artifact layer that lets engineers and researchers reproduce and compare results.

• Design for productization from day one (multi-tenancy, versioned APIs, security, and deployment in customer and air-gapped environments), and partner with Applied AI Research, product teams, and Dev Ops so the platform stays aligned with what they need.

Required Qualifications

• Production-grade Python and strong platform engineering fundamentals: API and service design, distributed systems, containers and Kubernetes, CI/CD, and testing.

• Shipped production agentic systems, with a clear sense of where they break and how to make them reliable.

• Experience with context engineering: retrieval-augmented generation, embeddings, vector or hybrid search, and memory for agents, ideally over code or large technical corpora.

• Experience instrumenting services (for example, with Open Telemetry…