Microsoft’s Frontier Tuning research team in Microsoft 365 is seeking Research Interns for summer 2027 in Redmond, Washington; the San Francisco Bay Area, California; or Baltimore, Maryland. The team aims to help organizations build AI that understands and works according to their knowledge, workflows, tools, and expert judgment. Its research uses reinforcement learning environments (RLEs) to let AI agents improve their skills, evaluations, harnesses, and underlying models while customer data remains within the customer’s trust boundary. The work supports Frontier Tuning in Microsoft Copilot Studio, Microsoft Foundry, and GitHub Copilot. The research addresses enterprise tasks that are long-horizon and open-ended, require expert-rubric evaluation, draw on varied information sources, and differ across organizations. The team has demonstrated that reinforcement fine-tuning in well-constructed enterprise environments can enable smaller tuned models to outperform frontier models on these tasks at lower cost. Researchers work with customers, product teams, and partners to bring methods to users and use field findings to guide research. Internship research may cover reinforcement learning and post-training for agents; enterprise environments and synthetic organizations; evaluation and rewards; learning from experts and product use; skills and organizational knowledge; and multimodal tuning. Interns collaborate with researchers, applied scientists, engineers, and product managers, and are encouraged to publish their work. The role is intended for PhD students interested in novel research grounded in practical needs and in seeing ideas reach products used by organizations worldwide.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
Job Type
Full-time
Career Level
Intern
Education Level
Ph.D. or professional degree