OpenAI
OpenAI
OpenAI is an artificial intelligence research and technology company building advanced AI systems with a focus on safety and responsible development. Its products include ChatGPT, which supports everyday and complex professional tasks, and an API platform that enables businesses and developers to integrate AI models into their own applications. Operating across artificial intelligence, SaaS, and enterprise technology, OpenAI combines research, product development, and efforts to improve how AI systems are governed and used.

Software Engineer, TPU Compilers

Improve inference performance for OpenAI’s frontier models on TPU hardware by profiling workloads, building kernels, and optimizing production execution.

Description

  • Optimize TPU model execution across the full forward pass, from model code and kernels to hardware behavior.
  • Profile workloads, find bottlenecks, and improve latency, throughput, and hardware utilization.
  • Build and tune kernels for model workloads.
  • Evaluate quantization methods that increase inference efficiency while maintaining model quality.
  • Create benchmarks and tools to measure performance, check correctness, and inform optimization work.
  • Work with researchers and engineers to support new model architectures and ship performance improvements.
  • Lead performance optimization across model code, kernels, and hardware execution.

Requirements

  • Hands-on experience optimizing machine learning workloads on TPUs for training, inference, or both.
  • Knowledge of modern model architectures and how their computations map to accelerator hardware.
  • Ability to relate kernel-level behavior to end-to-end model performance across software and hardware layers.
  • Strong software engineering capability.
  • A methodical approach to profiling, debugging, and validating performance gains.
  • Experience with quantization or attention kernels is an advantage.
  • Availability to work from the San Francisco office three days each week.
  • Authorization to work in the country where the role is based.

Benefits

  • Equity.
  • Performance-based bonuses for eligible employees.
  • Medical, dental, and vision coverage for employees and their families, with employer contributions to Health Savings Accounts.
  • Pre-tax Health FSA, Dependent Care FSA, and commuter accounts for parking and transit.
  • 401(k) plan with an employer match.
  • Paid parental leave of up to 24 weeks for birth parents and 20 weeks for non-birthing parents.
  • Up to eight weeks of paid medical and caregiver leave.
  • Flexible paid time off for exempt employees; non-exempt employees receive up to 15 days per year.
  • More than 13 paid company holidays.
  • Several coordinated paid company office closures during the year.
  • Paid sick or safe time.
  • Mental health and wellness support.
  • Employer-paid basic life and disability insurance.
  • Annual learning and development stipend.
  • Daily office meals and meal delivery credits, where eligible.
  • Relocation support for eligible employees.
  • Charitable donation matching and wellness stipends may also be available.

Related Jobs

Lambda

Senior Software Development Engineer, Network Automation

Lambda

Build automation for network provisioning, configuration, and capacity planning across Lambda’s AI cloud infrastructure. This hybrid role supports work from the Bellevue or San Francisco office.

Open