Yu Feng冯煜

I am an associate professor in the computer science department at UC Santa Barbara. I received my Ph.D. from UT Austin in 2018, where I was advised by Isil Dillig. My group builds automated reasoning tools at the intersection of programming languages, formal verification, and security — with recent focus on zero-knowledge proof systems and safe AI agents. Our work has been recognized with multiple Distinguished/Best Paper Awards at PLDI, CHI, and ASE, and has been supported by NSF, DARPA, Google, and the Ethereum Foundation.

My research develops principled techniques — rooted in program synthesis, type theory, and constraint solving — to automatically build, verify, and secure software systems. Current focus areas include: (1) formal methods and security for AI agents, ensuring autonomous systems behave safely and correctly through verification and runtime guarantees; (2) compilers and verifiers for zero-knowledge proof circuits, where we have built tools like Tabby and Tessel that detect vulnerabilities and optimize performance in ZK systems deployed in production; (3) smart contract and DeFi security, including attack synthesis and state-inconsistency detection; and (4) synthesis-driven tools for end-user programming, such as data visualization and code transpilation. A unifying theme is combining logical reasoning with learning to push the frontier of what automated tools can guarantee.