Shailja Thakur

I work on making agentic AI systems reliable enough for enterprise deployment. My current focus is on two problems: context optimization—how agents decide what to retain and discard across long, multi-step workflows—and automated debugging—systematically diagnosing why and where agent pipelines fail.

Before this, I pioneered the use of Large Language Models for automating hardware design (Verilog code generation, bug repair, security assertions)—work that received a Best Paper Award from ACM TODAES. My PhD at the University of Waterloo (with Sebastian Fischmeister) focused on security and interpretability in automotive systems. I did my PostDoc at NYU with Siddharth Garg and Ramesh Karri.