- San Francisco, California, United States
- On-site
- Full-time · Mid
Deeply technical Product Manager to make AI inference faster, smarter, and more efficient at scale, bridging systems engineering and real-world production workloads.
Compensation
Competitive compensation, including meaningful equity
Responsibilities
- Work with inference engineering and research teams to bring new performance capabilities to customers
- Translate advances in batching, caching, quantization, speculative decoding, and serving architectures into useful product features
- Define benchmarks and evaluations that demonstrate meaningful improvements in latency, throughput, quality, and cost
- Understand customer workloads and help prioritize the optimizations that will have the greatest impact
- Make performance features easier to configure, evaluate, and adopt in production
Requirements
- Nice to have: PyTorch
- Nice to have: vLLM
- Nice to have: TensorRT-LLM
- Nice to have: SGLang
- Nice to have: NVIDIA Dynamo