Join a pioneering team at Microsoft focused on the future of scalable software engineering. We're developing advanced AI agents capable of transforming massive codebases like Windows and Azure. Our work focuses on making large-scale software modifications trustworthy by integrating AI with static analysis, testing, and formal verification. This role offers a unique opportunity to blend cutting-edge research with high-impact production development, working closely with Microsoft Research and CoreAI.

Our current initiatives include migrating millions of lines of code across major Microsoft products to new versions and platforms, modernizing legacy C/C++ codebases to Rust with robust validation, and pioneering specification-driven development to infer formal specifications for existing code. These efforts aim to enhance code quality, discover subtle bugs, and establish validation guardrails for future AI-driven code changes.

This hybrid role requires at least three days a week onsite, fostering collaboration and continuous learning among talented colleagues. You'll be instrumental in shaping the next generation of software development tools.

Design, develop, and deploy agentic tools and infrastructure for critical areas such as large-scale migration, language modernization, and specification-driven development.

Construct the essential tool layers that AI agents rely on, including context, analysis, and validation mechanisms, alongside building evaluation harnesses to ensure functionality.

Integrate static analysis, testing, and verification techniques directly into agent workflows, guaranteeing that every code change is supported by empirical evidence.

Collaborate directly with engineering teams across Microsoft, including Windows and Azure, to successfully implement changes with production-level quality.

Implement instrumentation to meticulously measure quality metrics, such as acceptance rates and correctness, and actively iterate based on user feedback.

A strong foundation in Computer Science is essential, evidenced by a Bachelor's degree and 6+ years of experience, or a Master's degree with 4+ years, or a PhD. Proficiency in programming languages like C, C++, C#, or Rust is required, alongside demonstrated experience building developer tools leveraging LLM-based agents.

Expertise in compilers and program analysis is crucial. You should possess excellent technical design, problem-solving, and debugging capabilities. Strong communication and collaboration skills, a customer-centric mindset, and a commitment to delivering high-quality products are paramount.

Preferred qualifications include experience with Rust and C/C++ interop, memory safety, large-scale testing methodologies (e.g., differential testing, property-based testing, fuzzing), and formal methods. Familiarity with very large codebases and build systems, along with an interest in human-AI interaction and developer experience design, will be advantageous.