Research Engineer

Who you are

• PhD in CS/AI/ML (or equivalent research experience) with publications ideally in multi‑agent RL, agentic AI, or RL for language/code
• Strong Python and ML framework experience (PyTorch preferred; JAX/HF a plus)
• Demonstrated ability to turn research into working systems
• Experience designing evaluation environments and reward models for sequential/agentic tasks
• Experience and fluency with EDA tools (formal, simulation, physical design)
• Comfortable with data acquisition/curation; good instincts about data quality and licenses
• Clear communicator who partners well with other engineers
• Research on program synthesis/codegen, constrained decoding, or execution‑based rewards
• Experience with offline RL from tool traces or human corrections
• Open‑source contributions (e.g., SkyRL, verl, RLlib, Transformers, Pytorch)
• Familiarity with semiconductor/chip domains or other complex technical domains
• Track record of shipping research to production

What the job involves

• We’re hiring a Research Engineer to push the frontier of agentic LLMs and reinforcement learning for pushing the capabilities of Normal EDA, our agentic AI platform for semiconductor design automation. You’ll design and run experiments, build agents, curate datasets from complex technical artifacts, and create rigorous evaluations. You’ll write production‑quality research code and work closely with engineering to ship improvements to customers
• Build multi-agent systems for code generation that interact with EDA tools (e.g., simulations, waveform analysis, formal tools, physical design tools), propose fixes, and iterate through all stages of chip design and verification flows
• Build research prototypes that integrate with our production agentic code generation tool; collaborate to productionize wins
• Create RL environments and evaluations for agents, explore proxy rewards and consider speed/accuracy tradeoffs of custom tools
• Generate datasets from silicon collateral (e.g., RTL, testbenches, custom VIPs) sources such as RTL designs/VIPs/chip specifications/agent logs; generate synthetic data where appropriate; maintain data cards and licensing
• Analyze experiments with disciplined ablations; document results and drive progress with technical rigour
• Stay current on LLM agents, RL (offline/online, RLHF/RLAIF), constrained decoding, and program synthesis