Member of Technical Staff — Developer Technology

What you'll do

• Profile and optimize GPU workloads, diagnosing bottlenecks from kernels through distributed multi-node systems.
• Specialize in one or two areas, such as inference performance, kernels and hardware enablement, speculative decoding, or training systems.
• Work with ecosystem partners to deliver technical guidance and reproducible cookbooks, and feed improvements into SGLang, Miles, and the roadmap.

What they ask for

• 4+ years of experience in GPU systems, LLM infrastructure, or performance engineering.
• Strong profiling and debugging skills to diagnose performance and correctness issues across the stack.
• Hands-on GPU programming in at least one of CUDA, ROCm, or Triton, with willingness to work across platforms.
• Strong Python programming plus C++ or CUDA.

Nice to have

• Deep familiarity with LLM inference internals, low-precision quantization, custom GPU kernels, speculative decoding, or large-scale distributed training.
• Experience across NVIDIA and AMD platforms, or with SGLang, Miles, vLLM, TensorRT-LLM, Megatron, or comparable frameworks; open-source AI/ML contributions.

Pay

$200,000 - $400,000 USD annual base salary

Tools

SGLang, Miles, CUDA, ROCm, Triton, Python, C++, FP8

AI focus

Accelerate LLM inference and training on GPU hardware, including kernel optimization, model and hardware enablement, speculative decoding, and RL post-training.