Solving poker in custom WebGPU kernels
A developer recounts using AI coding assistants to build custom WebGPU kernels for a poker solver, bypassing the need for traditional tensor libraries. The project successfully achieved high performance in a browser environment by leveraging AI to generate and optimize low-level code.
Why it matters
This demonstrates the increasing capability of AI agents to assist in complex software engineering tasks, potentially reducing reliance on specialized third-party libraries.
Have coding agents gotten good enough that we don’t need libraries anymore? This post recounts one case in which the answer is “we don’t”: I needed a tensor library to run my poker model in WebGPU. The library I wanted did not exist. It turned out I did not need it.
For the past year, I’ve been interested in the state of solvers for poker. For those unfamiliar: a solver finds an approximate Nash equilibrium strategy for any game situation in poker. In practice, they take a “spot”, a set of public cards and betting history, and produce an output strategy. Because the strategy approximates an equilibrium, it is provably (up to epsilon) non-exploitable : if a player plays a different strategy than your equilibrium, they can’t beat you in expectation.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in