Software Engineer, TPU Compilers
Core
Lead the optimization of OpenAI's state-of-the-art frontier models on TPUs, owning performance across the full forward pass from model code through kernels to hardware execution.
Role type
Senior IC machine-learning systems engineer (TPU inference)
Builds
Optimized inference performance for OpenAI's AI models
Domain
AI/ML inference on Google TPU hardware
Deliverable
production ML models
Required skills
TPU optimization, kernel development, workload profiling, quantization, model architecture understanding, benchmarking, debugging
Preferred skills
experience with attention kernels, training optimization, system-level debugging
Technologies
TPUs, model kernels, quantization techniques
Responsibilities
Optimize model execution on TPUs, profile workloads to identify bottlenecks, develop and optimize kernels, build benchmarks and validation tools, partner with researchers on new model architectures
