Senior/Staff System Research Engineer – LLM Inference Optimization

Highlights

• Build next-generation intelligent inference systems for large-scale AI.
• Advance latency, throughput and cost efficiency for LLM serving.
• Drive research ideas from prototype to production-scale impact.

Ideal candidate

Independent systems thinker who thrives in fast-moving research environments. You love pushing performance frontiers and turning bold inference ideas into real production systems.

Requirements

• Experience in distributed AI systems, GPU systems, or high-performance computing
• Strong understanding of modern LLM inference architectures and performance tradeoffs in large-scale serving
• Hands-on experience with modern LLM inference and serving frameworks such as vLLM, SGLang, TensorRT-LLM or similar
• Experience designing, extending, or optimizing inference runtimes including scheduling, batching, KV-cache management and distributed execution
• Strong understanding of GPU architectures with CUDA, Triton or similar GPU programming environments
• Experience with CUTLASS, cuBLAS, cuDNN or related performance libraries
• Experience profiling end-to-end performance with Nsight Systems, Nsight Compute or equivalent tools
• Independent problem solver who defines technical questions and drives solutions through ambiguity
• Excellent communication and collaboration across research, engineering and product teams