Sachit Malik

Senior Security Engineer, Microsoft AI

Sachit Malik is a Senior Security Engineer specializing in AI and software supply chain security. He previously led several large-scale initiatives at Apple, including the design of enterprise-wide secure code generation frameworks, software supply chain and container security programs, and scalable secure-by-default tooling across thousands of repositories and microservices. His work integrates machine learning, formal methods, and program analysis to build trustworthy systems and evaluation and improvements in AI-generated code for security and compliance. Currently he is working on Security of, by, and for: LLMs and Agentic Systems (on domains such as: mitigating indirect prompt injection attacks in Agentic systems by design; detecting and mitigating control flow hijacking in multi-agent systems across long-horizon tasks; secure-by-default frameworks development (SecMCP SDK); and using non-traditional approaches based on model internals (RepE/MechInterp/Causal Tracing). for AI Safety and Security.)

Sachit holds an M.S. in Information Security, Policy, and Management from Carnegie Mellon University and a B.Tech. in Mathematics and Computing from IIT Delhi. He actively contributes to the AI Safety community through research in the domains of: LLMs and systems security, representation engineering, and agentic alignment, and serves as a reviewer for initiatives like the CSA Trusted AI Safety Expert (TAISE) program and a research fellow at the Supervised Program for Alignment Research (SPAR).