About this role
<b>Overview</b><br><p>Research Internships at Microsoft provide a dynamic environment for research careers with a network of world-class research labs led by globally-recognized scientists and engineers, who pursue innovation in a range of scientific and technical disciplines to help solve complex challenges in diverse fields, including computing, healthcare, economics, and the environment.</p><p> </p><p> </p><p>The Frontier Tuning research team in Microsoft 365 is seeking Research Interns for summer 2027 in Redmond, WA, the San Francisco Bay Area, CA, or Baltimore, MD. </p><p> </p><p>Our mission is to help every organization own its intelligence: Artificial Intelligence (AI) that knows, reasons, and works the way the organization does. We capture how an enterprise works, including its knowledge, sensemaking, workflows, tools, and expert judgment, in reinforcement learning environments (RLEs). These environments serve as a gym in which AI agents improve by tuning their skills, evaluations, harnesses, and underlying models, while customer data stays within the customer's trust boundary. Our research is the foundation of Frontier Tuning, announced at Microsoft Build 2026 and available to makers and developers in Microsoft Copilot Studio, Microsoft Foundry, and GitHub Copilot. </p><p> </p><p>Enterprise work is a demanding testbed for AI. Tasks are long-horizon and open-ended, quality is judged against expert rubrics rather than a single correct answer, knowledge is spread across documents, conversations, slides, and drawings, and every organization works differently. We have shown that reinforcement fine-tuning in well-constructed enterprise environments lets smaller tuned models outperform the latest frontier models on such tasks, at lower cost. We work side by side with customers, product teams, and partners so that new methods reach real users quickly, and the gaps we find in the field shape our next research questions. </p><p> </p><p>We are seeking Research Interns to advance the science of adapting AI agents to the enterprise. Research Internship projects will focus on topics such as: </p><ul><li>Reinforcement learning and post-training for agents, including RL for open-ended tasks without verifiable rewards, goal-driven training for long-horizon trajectories, tool use within training trajectories, harness-aware training, and efficient RL training and inference. </li><li>Enterprise environments and synthetic organizations, including scalable construction of realistic RL environments, simulation of multi-person collaboration, virtualized tools for side-effect-free training, and automatic generation of tasks, skills, and rubrics from enterprise signals. </li><li>Evaluation and rewards, including rubric-based grading, reliability of Large Language Models (LLM) judges, measuring environment quality, and evaluations that predict real-world value. </li><li>Learning from experts and from use, including learning from subject-matter-expert feedback, combining signals across the experts in a team, learning from sparse signals, and continual improvement after deployment. </li><li>Skills and organizational knowledge, including discovering (search), composing, and tuning agent skills, reasoning over organization-scale data, retrieval and embeddings, and deciding when knowledge belongs in context, in skills, or in model weights. </li><li>Multimodal tuning, including adapting models to slides, documents, and technical drawings, and to an organization's style and voice. </li></ul><p>This role is ideal for PhD students who want research that is both novel and grounded in real needs, and who want to see their ideas move into products used by organizations worldwide. You will collaborate with researchers, applied scientists, engineers, and product managers, and we encourage interns to publish their work. </p><br><br><b>Responsibilities</b><br><p>Research Interns put inquiry and theory into practice. Alongside fellow doctoral candidates and some of the world’s best researchers, Research Interns learn, collaborate, and network for life. Research Interns not only advance their own careers, but they also contribute to exciting research and development strides. During the 12-week internship, Research Interns are paired with mentors and expected to collaborate with other Research Interns and researchers, present findings, and contribute to the vibrant life of the community. Research internships are available in all areas of research, and are offered year-round, though they typically begin in the summer.</p><h4> </h4><h4>Additional Responsibilities</h4><ul><li>Design, implement, and evaluate new methods in our training, environment, and evaluation pipelines, together with researchers, applied scientists, and engineers. </li><li>Turn real enterprise scenarios into well-posed research questions, and compare against strong baselines, including the latest frontier models. </li><li>Document methods and findings, present results, demonstrate working prototypes, and prepare work for publication where appropriate. </li><li>Apply your knowledge, problem-solving skills, and drive to innovate to new challenges in the field, with the opportunity to see your ideas realized in products and services used worldwide. </li><li>Embody our culture and values. </li></ul><br><br><b>Qualifications</b><br><h4>Required Qualifications</h4><ul><li>Currently enrolled in a Doctorate Degree program in Computer Science, Electrical Engineering, or a related STEM field.</li></ul><p> </p><p> </p><p><strong>Other Requirements</strong></p><ul><li>Research Interns are expected to be physically located in their manager’s Microsoft worksite location for the duration of their internship.</li><li>In addition to the qualifications below, you’ll need to submit a minimum of two reference letters for this position as well as a cover letter and any relevant work or research samples. After you submit your application, a request for letters may be sent to your list of references on your behalf. Note that reference letters cannot be requested until after you have submitted your application, and furthermore, that they might not be automatically requested for all candidates. You may wish to alert your letter writers in advance, so they will be ready to submit your letter. </li></ul><p> </p><h4>Preferred Qualifications</h4><ul><li>Research experience in one or more of reinforcement learning, post-training of large language models, LLM agents and tool use, evaluation, natural language processing, information retrieval, or multimodal models, shown through publications or open-source work. </li><li>Hands-on experience training, fine-tuning, or post-training large language or multimodal models, for example with supervised fine-tuning, RL from human or AI feedback, preference optimization, or reinforcement fine-tuning. </li><li>Experience building environments, simulators, benchmarks, or evaluation methods for AI agents. </li><li>Demonstrated programming skills in Python, experience with deep learning frameworks such as PyTorch, and experience running large-scale experiments. </li><li>Ability to turn open-ended, real-world problems into well-posed research questions and rigorous experiments. </li><li>Ability to think unconventionally to derive creative and innovative solutions. </li><li>Ability to collaborate effectively with your mentor and others on new research projects, with strong communication and cross-group collaboration skills. </li></ul><p> </p><p> </p><div>The base pay range for this internship is USD $6,810 - $13,480 per month. There is a different range applicable to specific work locations, within the San Francisco Bay area and New York City metropolitan area, and the base pay range for this role in those locations is USD $8,920 - $14,610 per month.</div><div> </div><div> </div><div>Certain roles may be eligible for benefits and other compensation. Find additional benefits and pay information here: <a href="https://careers.microsoft.com/us/en/us-intern-pay" rel="noopener">https://careers.microsoft.com/us/en/us-intern-pay</a></div><div> </div><div> </div><div>Microsoft accepts applications and processes offers for these roles on an ongoing basis throughout the academic calendar (September - April).</div> <br><p>This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled.</p><br><hr><br><p>Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, citizenship, color, family or medical care leave, gender identity or expression, genetic information, immigration status, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran or military status, race, ethnicity, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable local laws, regulations and ordinances. If you need assistance with religious accommodations and/or a reasonable accommodation due to a disability during the application process, read more about <a href="https://careers.microsoft.com/v2/global/en/accessibility.html"><b><u>requesting accommodations.</u></b></a></p>