Join a small team at an early-stage AI infrastructure startup, building and operating systems that make large language model inference faster and more cost-efficient. The role combines low-level performance engineering, infrastructure operations, and customer problem-solving.
Bring new open-source models into production with a focus on latency and throughput.
Optimize inference systems, including batching, KV cache, speculative decoding, and quantization.
Write and tune kernels using CUDA, HIP, and Triton across a range of accelerator platforms.
Design, deploy, and operate heterogeneous clusters across infrastructure vendors.
Improve the reliability, observability, and cost efficiency of production inference fleets.
Work with customers to understand performance needs and help set technical priorities.
Typically 1 to 6 years of relevant experience, with strong new graduates also considered.
Strong computer science fundamentals, including understanding of computer architecture, memory hierarchy, and operating system internals.
Experience in backend, infrastructure, or systems engineering, rather than exclusively full-stack or web development.
Familiarity with CUDA, Triton, or GPU programming, and ideally experience with inference optimization or large-scale cluster operations.
Coursework or practical experience in operating systems, computer architecture, or compilers.
Evidence of exceptional technical work, strong communication skills, and experience using AI coding tools to improve engineering workflows.
Salary range: $200,000 to $300,000 annually. Visa sponsorship is available.
Remote within the United States. San Francisco, California is the primary location.
Job details are sourced from the employer's original posting.
Open job postingAbout the company
A talent solutions partner that helps companies build and scale their teams.