Everett Hildenbrandt
Everett Hildenbrandt has spent over 6 years leading the technical direction for RV's software tooling. He is passionate about providing high-quality and consistent developer tooling for all programming languages that emphasizes usability and power. In his journey from studying physics to working on validating the safety of distributed Web3 applications, he's seen that formal methods can play a crucial role in improving the quality of software for everyone. During his time at RV, he's driven broader adoption of formal verification through both education and bringing the verification tooling to the developers via improved UX.
Session
Everyone and their mother is trying to launch an AI auditing tool (usually meaning an LLM-based auditing tool). Do they perform any better than cats at auditing? Do they fill the same market purpose as auditing? This talk will take a praxeological approach to the question (as opposed to an empirical/scientific approach), so buckle up for some half-baked un-substantiated opinions! Mixed in will be some things I do think LLMs are good for in the auditing process, and some ideas of other approaches that may work better (some even being taken by some teams already).