OpenAISan Francisco, CAModels inference performance across application, model, and fleet layers using microbenchmarks to build cost-to-serve estimates. Analyzes workloads end-to-end, enhances bottleneck detection tools, and collaborates on optimizations for latency, throughput, and cost.
Improve agentic model capabilities, reliability, and product fit for power users and API developers through evals, training data, and post-training interventions.
Build and maintain infrastructure for large-scale RL training runs of frontier OpenAI models. Debug across training, inference, and distributed systems while supporting research integrations.
Hands-on senior software engineer focused on maintaining and improving ML training infrastructure, debugging training systems, and unblocking researchers on the robotics team.
Designs and scales distributed data infrastructure for large-scale multimodal AI training and evaluation. Collaborates with researchers to build reliable, high-performance systems in a fast-paced environment.
Develops and optimizes OpenAI's inference infrastructure for AMD GPUs, handling low-level kernel performance, distributed execution, and integration with serving frameworks like vLLM and Triton. Requires expertise in GPU programming with HIP/CUDA and distributed systems scaling.