Research Engineer, Agentic EDA

Job Title

Research Engineer, Agentic EDA

Role Summary

Develop and evaluate agentic large-language-model (LLM) systems and reinforcement-learning methods to advance Normal's agentic EDA platform for semiconductor design automation. Work spans research experiments, dataset curation from silicon artifacts, building prototypes that integrate with production systems, and creating rigorous evaluations.

Collaborate with engineering teams to productionize research outcomes. The team operates across multiple offices including New York City, Mountain View (Silicon Valley), London, Copenhagen, and Zurich.

Experience Level

Mid-level (Level - Mid-Career). No explicit years-of-experience requirement stated.

Responsibilities

Key responsibilities include designing experiments, building agents and evaluation environments, and producing research that can be deployed.

• Design, implement, and run experiments for multi-agent systems that generate code and interact with EDA tools (simulators, waveform analysis, formal tools, physical design tools) across chip design and verification flows.
• Build research prototypes that integrate with production agentic code-generation tools and collaborate to productionize successful approaches.
• Create RL environments, define proxy rewards, and evaluate trade-offs between speed and accuracy for tool-driven agents.
• Generate and curate datasets from silicon collateral (RTL, testbenches, VIPs, chip specifications, agent logs) and produce synthetic data where appropriate; maintain data documentation and licensing records.
• Analyze experimental results with disciplined ablations and document findings to drive technical progress.
• Maintain currency with advances in LLM agents, RL (offline/online, RLHF/RLAIF), constrained decoding, and program synthesis.

Requirements

Must-have technical skills and experience; followed by concise nice-to-have items.

• Must-have: Strong Python skills and experience with ML frameworks (PyTorch preferred; JAX/Hugging Face a plus).
• Proven ability to turn research ideas into working systems and collaborate with engineering teams to ship improvements.
• Experience designing evaluation environments and reward models for sequential or agentic tasks.
• Fluency with EDA tools and workflows (formal verification, simulation, physical design).
• Experience acquiring, curating, and assessing data quality and licensing for technical datasets.
• Clear written and verbal communication; ability to work effectively with cross-functional engineers.
• Nice-to-have: Research or engineering experience in program synthesis/code generation, constrained decoding, execution-based rewards, offline RL from tool traces or human corrections, and open-source contributions in related projects.

Education Requirements

PhD in Computer Science, AI, Machine Learning, or a related field preferred; or equivalent research experience. Publications in multi-agent RL, agentic AI, or RL for language/code are noted as ideal. The posting allows equivalent practical or research experience in lieu of a degree.

About the Company

Company: Normal Computing

Headquarters: New York, NY, USA

Normal Computing develops software and hardware solutions for the semiconductor and AI infrastructure industries, specializing in ASICs for image and video diffusion inference and AI accelerators. The company focuses on architecture and microarchitecture of compute blocks, PE array design, ISA co-design, and FPGA prototyping.