RL Research Engineer for Chip-Design AI Tools

OpenAI; San Francisco (CA); On-site; USD 140,000 - 210,000; Full time

Job summary

OpenAI is hiring a Research Engineer to advance chip-design problem solving using reinforcement learning, tool use, and evaluation. You will own experiments from idea to implementation and analysis, building environments and evaluations, running training, and ensuring reproducible results with strong coding practices.

Experience in reinforcement learning, model evaluations, tool-using agents, and testing hypotheses will help you thrive.

Qualifications

• Proven ability to turn technical ideas into working software.
• Experience in reinforcement learning, model evaluations or post-training analysis.
• Experience building tool-using agents or automated evaluation systems.
• Ability to form clear hypotheses and design useful experiments.
• Able to work independently on ambiguous problems and explain results clearly.

Responsibilities

• Build RL environments and evaluations for RTL generation, design verification, and physical design optimization.
• Develop and test methods to help models use chip-design tools and improve power, performance, and area while preserving correctness.
• Design experiments, establish baselines, and verify improvements on new tasks and designs.
• Investigate failures across model behavior, rewards, evaluation tools, and infrastructure.
• Turn successful experiments into reusable research code and training workflows with researchers and engineers.

Skills

Strong programming & debugging

Reinforcement learning experience

Experiment design

Independent work

Cross-team communication

Tools

RTL/Verilog/SystemVerilog

EDA tools

Experiment orchestration

Distributed training

Job description

OpenAI is hiring a Research Engineer to advance chip-design problem solving using reinforcement learning, tool use, and evaluation. You will own experiments from idea to implementation and analysis, building environments and evaluations, running training, and ensuring reproducible results with strong coding practices.

Experience in reinforcement learning, model evaluations, tool-using agents, and testing hypotheses will help you thrive.