Member of Technical Staff — DevOps

Description

We look for infrastructure engineers who are obsessed with engineering velocity. Everything we do — research, training, serving, customer deployments — moves only as fast as our tooling. Your mission is to design, build, and operate the platform underneath it all, including the internal infrastructure for agentic development, so every engineer and researcher can iterate rapidly with agents doing more of the work.

Responsibilities

• Measure developer velocity, and partner with researchers, product engineers, and forward-deployed engineers to find what slows them down and remove it
• Build internal infrastructure for agentic development: sandboxes, tool access, context, and eval harnesses that let autonomous research and coding agents run end to end
• Self-host and serve open-source LLMs for internal agent use
• Own CI/CD, build and test infrastructure, reproducible dev environments, and infrastructure-as-code so research and product code ship quickly and safely
• Build the monitoring, alerting, and on-call for our product and customer-facing services to catch failures before users do
• Own cost across the dev stack: track and cut spend on compute, tools, and inference

Requirements

• We value a relentless approach to problem-solving, rapid execution, and the ability to quickly learn in unfamiliar domains.
• AI-pilled: uses coding agents daily and has built internal infrastructure for agentic development, not just used it
• Experience self-hosting and serving open-source LLMs (e.g. vLLM, SGLang)
• Strong systems background: Linux, networking, containers and orchestration (e.g. Docker, Kubernetes), infrastructure-as-code (e.g. Terraform)
• Experience building CI/CD and developer tooling (e.g. GitHub Actions, Buildkite, Bazel, Nix)
• Knowledge of cloud platforms (GCP, AWS, or Azure) and best practices for observability and security
• Track record of measurably improving developer velocity
• Owns deliverables end-to-end, from requirements through autonomous execution