<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Chelis blog</title><description>Notes from the people building Chelis.</description><link>https://chelis.ch/</link><item><title>Does proof beat fuzzing? Testing Chelis on FPBench</title><link>https://chelis.ch/blog/proof-vs-fuzzing/</link><guid isPermaLink="true">https://chelis.ch/blog/proof-vs-fuzzing/</guid><description>We planted bugs in 103 FPBench kernels and compared Chelis&apos; proof mode with fuzzing alone, at default settings and at equal sampling budgets.</description><pubDate>Tue, 27 Oct 2026 00:00:00 GMT</pubDate><author>Gerasimos Lampouras</author></item><item><title>Chelis: A Tensor Language Agents Can Write and Prove</title><link>https://chelis.ch/blog/launch/</link><guid isPermaLink="true">https://chelis.ch/blog/launch/</guid><description>Chelis is a language for numerical code written by AI coding agents, with shapes and precision in its types and a prover built in.</description><pubDate>Tue, 06 Oct 2026 00:00:00 GMT</pubDate><author>Jeff Smith</author></item></channel></rss>