AI Research Scientist

Cerebro; San Francisco (CA); On-site; USD 150,000 - 230,000; Full time

Job summary

Cerebro in San Francisco is seeking a PhD or MS-level researcher/engineer to advance agentic AI for advanced semiconductor design. You will work across research and engineering to develop AI and agentic methods, build prototypes and deploy techniques into production environments.

Join a deeply technical team applying LLMs, reinforcement learning, and GPU-accelerated computing to design automation in EDA. Strong implementation skills and publishing or production experience are valued.

Qualifications

• PhD or Master’s degree in Computer Science, Electrical Engineering, Computer Engineering or related field, or equivalent practical experience.
• Strong Python skills and proficiency in C++ or Rust.
• Experience with PyTorch, JAX or comparable ML framework.

Responsibilities

• Develop AI and agentic methods for semiconductor design and verification.
• Build agents capable of understanding engineering objectives, using EDA tools, executing multi-step workflows and recovering from failures.
• Create tools and representations that allow autonomous systems to reason over design states, constraints, actions and optimization objectives.
• Research and implement methods involving LLMs, reinforcement learning, program synthesis, search and combinatorial optimization.
• Develop GPU-accelerated algorithms for computationally intensive design, simulation and analysis problems.
• Design rigorous evaluations for engineering agents operating with private, sparse or customer-specific data.
• Integrate AI systems with simulators, formal tools, design databases, commercial EDA software and customer infrastructure.
• Work directly with semiconductor engineers to identify high-value automation opportunities.
• Translate promising research into reliable and scalable product capabilities.

Skills

Python

C++

Rust

PyTorch/JAX

Agentic AI

LLMs

Research

Education

Master's or PhD in CS/EE/related field

Tools

CUDA

Triton

Verilog/SystemVerilog

GPU programming

Job description

We are partnering with an early-stage technology company building a new generation of AI-driven design automation for the semiconductor industry.

The team brings together expertise across artificial intelligence, electronic design automation, semiconductor engineering, GPU computing and production software. Its platform combines autonomous agents, specialised engineering tools and continuous learning to automate complex chip-design workflows.

This is an opportunity to join a small, deeply technical team and help define how agentic AI will be applied to real semiconductor design.

The Role

You will develop new technologies at the intersection of agentic AI, machine learning, GPU-accelerated computing and electronic design automation.

Working across research and engineering, you will identify important technical problems, develop new algorithms and agent-native tools, build working prototypes and deploy successful approaches into real semiconductor environments.

The role is particularly suited to someone who combines strong research ability with exceptional implementation skills and wants to see their work progress beyond papers and prototypes into production.

Responsibilities

Develop AI and agentic methods for semiconductor design and verification.

Build agents capable of understanding engineering objectives, using EDA tools, executing multi-step workflows and recovering from failures.

Create tools and representations that allow autonomous systems to reason over design states, constraints, actions and optimisation objectives.

Research and implement methods involving LLMs, reinforcement learning, program synthesis, search and combinatorial optimisation.

Develop GPU-accelerated algorithms for computationally intensive design, simulation and analysis problems.

Design rigorous evaluations for engineering agents operating with private, sparse or customer-specific data.

Integrate AI systems with simulators, formal tools, design databases, commercial EDA software and customer infrastructure.

Work directly with semiconductor engineers to identify high-value automation opportunities.

Translate promising research into reliable and scalable product capabilities.

Requirements

A PhD or master’s degree in Computer Science, Electrical Engineering, Computer Engineering or a related field, or equivalent practical experience.

Strong Python skills and proficiency in a systems language such as C++ or Rust.

Experience with PyTorch, JAX or a comparable machine learning framework.

Research or engineering experience in at least one of the following areas:

Agentic AI or large language models

GPU-accelerated or parallel computing

Program synthesis or code generation

Formal methods

Machine learning for scientific or engineering applications

Particularly Valuable Experience

Publications in leading AI, EDA, systems, HPC or computer architecture venues.

Developing EDA algorithms, optimisation engines or domain-specific engineering tools.

GPU programming with CUDA, Triton or related technologies.

Functional verification, RTL development, synthesis, timing analysis, physical design or analog design.

Building agents that interact with tools, codebases, databases or external environments.

LLM training, post-training, fine-tuning, retrieval, tool use or evaluation.

Familiarity with Verilog, SystemVerilog, SPICE, TCL or commercial EDA tools.

Why Join?

Help define an emerging category at the intersection of AI and semiconductor design.

Build autonomous systems that perform complex engineering work rather than simply generate recommendations.

See your research deployed within real chip-design organisations.

Have meaningful influence over both the research direction and the resulting product.