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.
We are seeking multiple Research Interns to work with researchers in Microsoft Research Special Projects and the Deep Learning Group, alongside partner product teams, to advance core AI research across security, agentic workflows, and next‑generation architectures.
1. Safety and Security
Develop methods to strengthen the robustness and trustworthiness of AI systems. This includes defenses against data poisoning, jailbreak attacks, and other adversarial threats, as well as principled threat evaluation and risk modeling through statistical analysis and the establishment of realistic attack benchmarks.
2. Post-Training Alignment and Multi-Turn Interaction
Design alignment strategies and fine-tuning harnesses that enable post-training for efficient, coherent multi-turn interactions in agentic systems to ensuring aligned models maintain guardrails under complex tool use and orchestration.
3. Next-Generation Architectures
Investigate architectures beyond transformers to address task-specific needs, such as models for controlled dynamical systems, interpretable designs inspired by neuroscience, and multi-modal integration for richer reasoning.
Projects span fundamental research and system-level applications, leveraging shared datasets, training and compute resources, and unified evaluation frameworks to enable impactful cross-domain applications.