Exploring tools and approaches that make us more effective engineers and make our systems safe and reliable. Join us for discussions on recent developments in topics such as best practices in reliability and security, applied formal methods, encryption, and safe and secure hardware.
All content for Building Better Systems is the property of Galois, Joey Dodds, Shpat Morina and is served directly from their servers
with no modification, redirects, or rehosting. The podcast is not affiliated with or endorsed by Podjoint in any way.
Exploring tools and approaches that make us more effective engineers and make our systems safe and reliable. Join us for discussions on recent developments in topics such as best practices in reliability and security, applied formal methods, encryption, and safe and secure hardware.
#7: Aditya Thakur – “If it goes too slow, they'll turn it off”: Analysis Tools That Work
Building Better Systems
1 hour 13 minutes 38 seconds
4 years ago
#7: Aditya Thakur – “If it goes too slow, they'll turn it off”: Analysis Tools That Work
Dr. Aditya Thakur, a computer science professor at U.C. Davis, walks us through his work on developing analysis tools that he wished he had while working in industry at places like Google. Aside from program analysis, we talk about making a research group successful by exposing them to industry. Towards the end, he shares his work on techniques and tools for repairing a trained deep neural network once a mistake has been discovered. Along the way, we learn about things like abstract interpretation, non-determinism, the trickiness of parallelism, and other concepts pertinent to analysis in an approachable way.
Building Better Systems
Exploring tools and approaches that make us more effective engineers and make our systems safe and reliable. Join us for discussions on recent developments in topics such as best practices in reliability and security, applied formal methods, encryption, and safe and secure hardware.
