Onur Kasimlar
AI Meets Hoare Logic: Revolutionizing Software Testing with Formal Methods
#1about 3 minutes
The case for software verification in the AI era
AI-generated code is probabilistic and requires more than traditional testing, creating a need for formal software verification to ensure correctness.
#2about 3 minutes
Understanding the difference between testing and verification
Software testing finds errors through execution with selected inputs, while formal verification provides a complete mathematical proof of correctness for all inputs.
#3about 6 minutes
A primer on Hoare Logic and its core axioms
Hoare Logic uses preconditions, postconditions, and axioms like assignment and composition to formally prove that a piece of code behaves as specified.
#4about 1 minute
Current tools and applications for software verification
Modern verification relies on tools like SMT solvers and model checkers and is primarily used in critical systems like aerospace due to its high manual effort.
#5about 4 minutes
Verifying a binary search algorithm with OpenJML
A practical demonstration shows how to use OpenJML annotations like 'requires' and 'ensures' to formally verify a Java binary search implementation.
#6about 3 minutes
Using AI to generate formal verification proofs
AI tools like Gemini can generate verification annotations but may introduce subtle errors, highlighting the need for human oversight in this new validation workflow.
Related jobs
Jobs that call for the skills explored in this talk.
Matching moments
01:07 MIN
Defining formal methods for software verification
When testing just doesn’t cut it
04:36 MIN
How formal verification proves code correctness
When testing just doesn’t cut it
06:52 MIN
How AI will reshape software development and documentation
Coffee with Developers - Scott Chacon on growing GitButler and the future of version control
03:26 MIN
Envisioning the future of testing with artificial intelligence
How will artificial intelligence change the future of software testing?
08:56 MIN
The need for human expertise in the era of AI code
WeAreDevelopers LIVE - Accessibility isn't magic, Longevity, Devrel in times of AI and more
02:27 MIN
Maintaining code quality with AI-generated code
The AI-Ready Stack: Rethinking the Engineering Org of the Future
02:47 MIN
Final perspectives on the future of AI in software
From Monolith Tinkering to Modern Software Development
00:48 MIN
Building human-in-the-loop systems with traceable AI
The shadows of reasoning – new design paradigms for a gen AI world
Featured Partners
Related Videos
27:54When testing just doesn’t cut it
Lars Hupel
45:24How will artificial intelligence change the future of software testing?
Evelyn Haslinger
25:14The shadows of reasoning – new design paradigms for a gen AI world
Jonas Andrulis
28:09The AI-Ready Stack: Rethinking the Engineering Org of the Future
Jan Oberhauser, Mirko Novakovic, Alex Laubscher & Keno Dreßel
15:26New AI-Centric SDLC: Rethinking Software Development with Knowledge Graphs
Gregor Schumacher, Sujay Joshy & Marcel Gocke
04:17Leapter: The Reinvention of Software Development? A Future Built On AI Generated Code.
Robert Werner
31:56Collaborative Intelligence: The Human & AI Partnership
Prashanth Chandrasekar, Alejandro Saucedo, Jakob von Lindern & Demetris Cheatham
28:27AI & Ethics
PJ Hagerty
Related Articles
View all articles
From learning to earning
Jobs that call for the skills explored in this talk.
Think-cell
Foster City, United States of America
Intermediate
Unit Testing
Manual Testing
Automated Testing
webLyzard
Vienna, Austria
DevOps
Docker
PostgreSQL
Kubernetes
Elasticsearch
+2
Government of The United Kingdom
Amdocs
Kontich, Belgium
Senior
Terraform
Kubernetes
Machine Learning
Continuous Integration
Infosupport
Veenendaal, Netherlands
€0K
Natural Language Processing