About the role
Inferact's mission is to grow vLLM as the world's AI inference engine and accelerate AI progress by making inference efficient and faster. Founded by the creators and core maintainers of vLLM, we sit at the intersection of models and hardware—a position that took years to build.
About
the Role vLLM is growing at a fast pace, and every bit of that growth lands on the CI system. More models, more hardware, more contributors, more ways for things to break. Your job is to advance the CI system so it scales with vLLM’s momentum and unlocks faster development for everyone. You’ll get to: Maintain and scale the compute infrastructure that powers CI, release, performance benchmark, accuracy evaluation for vLLM project, across a wide range of models and accelerators including H100/H200, (G)B200/300, AMD MI325/355X, TPU, Intel Gaudi, etc.. Get creative about cutting CI time-to-signal from hours to minutes Make sure every corner of vLLM code base is well-tested Keep vLLM releases rock-solid Build out tooling that helps 3,000+ vLLM contributors move fast Skills and
Qualifications Minimum
qualifications: Strong experience with Docker, Kubernetes, and containerized build or test environments. Built CI/CD pipelines from scratch using GitHub Actions, Buildkite, or similar systems. Familiar with CI design patterns and CI techniques: compute orchestration, handling flaky tests, dependency/environment management, caching, remote execution, test target determination, etc, test coverage, and so on. Fluent in Python, Bash, Go, or similar for automation and tooling. Solid fundamentals of Linux, security, networking, storage, package management,.
Bonus points for: Setting up infrastructure for ML, inference, CUDA, ROCm, or accelerator-heavy workloads. Running Buildkite at scale, including agents, queues, dynamic pipelines, test sharding, caching, and artifact management. Operating Kubernetes clusters for CI, batch jobs, test execution, or internal developer infrastructure. Managing CI/CD in large open-source project Building dashboards, alerts, runbooks, or tooling for CI observability. Logistics Location: This role is based in San Francisco, California. Will consider remote in the US for exceptional candidates.