Theorem Launches: AI Coding That Is Trustworthy-By-Default

Theorum recently launched!

‍

^{"AI-coding IDE for mission-critical software"}

‍

^{TL;DR: Please give them your most complicated coding problem that needs to be correct. Sign up!}

‍

Founded by Jason Gross & Rajashree Agrawal

‍

MISSION

Theorem is an AI and programming languages research lab. Their question: as the models get better at generating code, what is the bottleneck to safely deploying vastly more computation in the world?

• Code has bugs.  AI-enabled attackers will increase the volume and velocity of attacks on software infrastructure. Human code review will not scale.
• AIs are untrusted generators. In order to deploy AIs in critical systems without fully understanding their generalization behavior, they wull need robust, scalable methods for overseeing their solutions.

‍

OPPORTUNITY

Programming languages and formal methods is the promise that they can bootstrap roots of trust. Even as complexity of software goes up, they can design simple interfaces for steering with leverage.

Previously, this technology was only applied to the most mission-critical infrastructure, because of the enormous labor bottleneck — requiring years of PhD level engineering work. For instance, Jason worked on the project that wrote the code generator powering trillions of secure https connections everyday, but it took fifteen man-years to complete!

With advances in AI, they are able to speed this development cycle 10,000 times! As a wild example of what’s possible, they spent two hours finding what would have been a zero-day in https code from 2011.

‍

THEIR APPROACH

The team wants to build workflows for trustworthy-by-default programming by pushing their best ideas till they break. The current candidate to beat is program equivalence driven development. You oversee a simple reference implementation. They give you an optimized implementation, along with a proof that the programs are functionally equivalent.

‍

Theorum is gradually rolling out their programming environment and models. Some use cases from their customers:

• Performance optimization: finding edge cases where hyper-optimized GPU code would fail; getting clear documentation of dependencies that modify behavior in unforeseen ways.
• Code migration: automating legacy code refactors from Python to Rust while proving no functional change; writing custom kernels to support new services.

‍

Learn More

‍

^{🌐 Visit theoremlabs.com to learn more.}

^‍

^{🤝 If you work in public infrastructure, finance, hardware, or have other use cases that demand correctness, Theorum would love to help. Please reach out to the founders here.}

‍

^{⭐ Give Theorum a star on Github.}

‍

^{👣 Follow Theorum on LinkedIn & X.}

‍