Senior Researcher - Security & AI

Responsibilities 

• Lead research at the intersection of security and AI, spanning both securing AI systems and applying AI to security problems. 

• Develop new techniques to protect sensitive and proprietary data in AI systems and pipelines, including training, post-training, reinforcement learning, inference, and agentic workflows. 

• Investigate the security of AI agents and autonomous systems, including attacks arising from tool use, code execution, memory, external data, multi-agent interaction, and long-running autonomous behavior, and develop principled defenses against them. 

• Explore how AI can help address security problems, including discovering vulnerabilities and attack paths, understanding complex security environments, detecting attacks and anomalous behavior, and developing effective defenses. 

• Investigate new attacks enabled or amplified by advances in AI and develop robust defenses against them. 

• Design rigorous threat models, benchmarks, evaluations, and experimental methodologies for emerging AI security problems. 

• Form hypotheses, build prototypes, design experiments, and evaluate new security mechanisms in realistic environments. 

• Pursue  long-term research while identifying opportunities to translate promising ideas into practical security mechanisms. 

• Publish at top-tier security, systems, cryptography, and machine learning venues, release research artifacts where appropriate, and help shape the research agenda in security and AI. 

• Collaborate with research and product teams across Microsoft to translate research advances into stronger security for real-world Microsoft systems. 

• Mentor interns and entry level researchers and contribute to MSR India's broader academic and research engagement. 

• Embody our culture and values.  

Required Qualifications  

• PhD in Computer Science, Computer Engineering, Mathematics, or a related technical field, or equivalent experience. 

• 1+ years of software engineering or research experience. 

• Strong background in one or more areas of computer security, privacy, cryptography, computer systems, or machine learning. 

• Strong research track record in security, cryptography, systems, machine learning, or a closely related area. 

• Demonstrated ability to conduct independent research and communicate results through publications, prototypes, open-source artifacts, or other impactful outcomes. 

• Ability to identify and formulate important research problems, develop novel technical approaches, and evaluate them rigorously. 

• Ability to work effectively across disciplinary boundaries and collaborate with diverse research and engineering teams. 

Preferred Qualifications  

• Research experience in AI security, security of LLMs, or security of agentic systems. 

• Experience with security and privacy techniques for AI training, post-training, inference, or data pipelines. 

• Experience in areas such as systems security, applied cryptography, privacy-enhancing technologies, confidential computing, access control, or secure machine learning. 

• Experience studying adversarial behavior, software vulnerabilities, attack surfaces, or security of large-scale systems. 

• Experience applying LLMs or other AI techniques to security problems such as vulnerability discovery, security analysis, attack detection, or defense generation. 

• Experience building and evaluating large-scale security or AI systems. 

• Experience designing threat models, security benchmarks, adversarial evaluations, or red-team methodologies for emerging systems. 

• Experience working across the research-to-production boundary and translating research ideas into real-world systems. 

How to Apply  

Please apply using the link above. In addition, please send an email to the Hiring Committee at seas-hiring-msri@microsoft.com.