Member of Technical Staff - GPU Inference Engineer

Liquid AI•San Francisco, CA
•Hybrid

About The Position

Liquid AI is hiring a Member of Technical Staff to improve its GPU serving stack, built on vLLM and SGLang. The role focuses on making Liquid Foundation Models faster and cheaper to serve for evaluation runs, synthetic data generation, and the company’s API service. It is a high-ownership position in the Research & Engineering team, with work focused at the serving-framework level rather than on custom CUDA kernels. The engineer will identify and address inference bottlenecks, profile workloads, and carry improvements through to benchmarked results. They will collaborate with researchers and serving engineers, design inference pipelines for large-scale evaluation and synthetic data generation, select and tune the inference engine and configuration behind the API service, and integrate Liquid Foundation Models into public serving frameworks and libraries.

Requirements

  • Hands-on experience running LLM inference on GPUs with a serving framework such as vLLM or SGLang, beyond default configuration.
  • First-principles understanding of LLM inference, including the transformer forward pass, autoregressive decoding on a GPU, and resulting latency, throughput, and memory trade-offs.
  • Solid understanding of GPU fundamentals as they apply to inference.
  • Experience profiling and benchmarking inference workloads and acting on the results.
  • Strong Python and PyTorch skills.
  • Comfort reading and changing large open-source codebases.

Nice To Haves

  • Merged contributions to vLLM, SGLang, or FlashInfer.
  • Lower-level GPU optimization experience, such as CUDA or Triton kernels.
  • Experience building high-throughput batch inference for evaluation or synthetic data generation.

Responsibilities

  • Own and improve the GPU serving stack on vLLM and SGLang, including configuration, scheduling choices, and framework changes.
  • Speed up large-scale evaluation and synthetic data generation, and design the inference pipelines used for those workloads.
  • Select, profile, and benchmark the inference engine behind the API service, and adapt the API layer around it.
  • Integrate Liquid Foundation Models into vLLM, SGLang, and Hugging Face Transformers, and maintain the public integrations as models and frameworks change.
  • Diagnose throughput and latency problems with researchers and serving engineers.
  • Profile and benchmark inference workloads, then use the results to identify and implement improvements.

Benefits

  • Competitive base salary plus equity.
  • Unlimited paid time off.
  • Company-wide Refill Days and long weekends.
  • Comprehensive health coverage, with details varying by location.
  • Paid leave for pregnancy, parenting, medical needs, and caring for family.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service