Client Roles
ABOUT THE ROLE
Join an early-stage AI infrastructure team building systems that improve the performance and efficiency of open-source model inference. This role spans low-level optimization and production operations, helping deliver fast, reliable inference across diverse hardware.
WHAT YOU'LL DO
- Bring new open-source models into production with optimizations for latency and throughput.
- Improve inference serving through batching, KV cache management, speculative decoding, and quantization.
- Write and tune GPU kernels using CUDA, HIP, and Triton across a range of accelerator platforms.
- Design, deploy, and operate heterogeneous clusters across infrastructure providers.
- Run production inference across mixed hardware fleets, improving reliability, observability, and cost efficiency.
- Work with customers to understand needs and solve infrastructure performance problems.
WHAT WE'RE LOOKING FOR
- Approximately 1 to 6 years of experience in backend, infrastructure, or systems engineering, including low-level work beyond application development; strong early-career candidates are welcome.
- Understanding of computer architecture, memory hierarchies, and operating system internals.
- Experience operating infrastructure at scale, such as clusters, data centers, or GPU fleets.
- Familiarity with machine learning inference optimization, including quantization, batching, KV caches, or speculative decoding.
- Experience with CUDA, Triton, or other GPU programming, with Python and exposure to technologies such as HIP, TPU, or AWS Trainium useful.
- St