My work focuses on the implementation and application of a novel tool called Mimicry Monitors. 

What are Mimicry Monitors?

They are a program verification technique that leverages shared code between program versions to reduce the computational cost of analysis. Given two programs that share a common fragment, a Mimicry Monitor is constructed to simulate the behavior of one program using the other as a reference, exploiting structural similarities to avoid redundant verification work. This approach integrates naturally into regression testing and fuzzing pipelines, where programs evolve incrementally and re-analyzing unchanged code from scratch represents a significant overhead. The result is a more scalable verification workflow that preserves formal guarantees while reducing the resources required to achieve them.