Abstract: Runtime verification promises to make software more reliable by monitoring program executions against formal specifications of safety properties. This talk will present results from work that my group did over the last several years to realize this promise in the context of software testing. I will start by briefly summarizing our early results showing that runtime verification during testing finds hundreds of confirmed bugs that testing alone misses in large open-source software, and that exploiting developers' small but frequent code changes scales runtime verification better during continuous integration. Next, I will discuss four challenges that we identified, which must be addressed if runtime verification is to become more widely used among developers. Lastly, I will present our recent results and works-in-progress that address these challenges, and (if time permits) new types of software tests that we invented in the process.

Short Bio: Owolabi Legunsen is an assistant professor in the Department of Computer Science at Cornell University. His research is on software engineering, with a focus on software testing, runtime verification, and the unification of both approaches. Owolabi's work has been recognized with four ACM SIGSOFT Distinguished Paper awards, an NSF CAREER Award, and an Intel Rising Star Faculty Award.