About this role
Responsibilities
- Define, document, and maintain best practices for AI-assisted coding, testing, code review, and research workflows.
- Influence guidelines for using LLMs on performance-critical software, including embedded C++ and Rust, RTOS applications, and kernel device drivers: where AI assistance is appropriate, how generated code is reviewed and tested, and how memory safety, timing, and determinism are verified before merge.
- Make forward-looking electronics architecture improvements to our AI tooling and integrations so they keep pace with advances in models and agents and with shifts in engineers' tool preferences, and track industry benchmarks for AI-assisted developer productivity so our practices stay current.
- Build ways to measure the output of AI and LLM-driven workflows, such as defect and revert rates of AI-assisted changes, review findings, test coverage, performance regressions, new capabilities added, and time saved, and use the results to refine tools and guidelines.
- Ensure safe and reliable operation of AI in quantum processors.
Experience & Education
- Master’s or PhD in Computer Science, Software Engineering, or Artificial Intelligence
- 8+ years of software development
Qualifications
- Strong software engineering skills in Python, and experience writing embedded software in C++ or Rust, including manual memory management, hard real-time constraints, and kernel device driver implementation.
- Influencing an organization of > 50 engineers using AI agents (Claude Code, Cursor, or GitHub Copilot) on a daily basis, identifying pitfalls and measuring impact.
- Working knowledge of how LLMs behave in practice: context management, prompting, agentic workflows, harnesses, effective safety practices, tool use, and judging output quality.
- Clear written and spoken communication with audiences ranging from physicists to new software hires. An ability to influence a multi-discipline engineering organization.