Jesse Alama

Jesse Alama

@jessealama.net

Math and programming. Lean, JS, Racket. American. Igalian.

183 followers60 following174 posts

Jesse Alama reposted

I love formal verification + fuzzing techniques but am wary that we're quickly speedrunning towards a future where people are generating enormous incomprehensible correctness proofs about their programs, but are none-the-wiser on what properties have actually been proven.

Joe Fabisevich@mergesort.me

I don't say this lightly: this approach is clearly the future of software development. I have a blog post brewing that touches on the subject, but fuzz testing and formal verification at scale will make all software an order of magnitude better, safer, and more reliable.

I'm happy to announce pabst: a tool for Property-Based Testing in JavaScript/TypeScript. Annotate your functions with properties they're supposed to have, run `pabst test` to try to invalidate those claims, and watch the bugs come crawling out. Install via `npm install pabst-checker`,

View full thread

I'm happy to announce pabst: a tool for Property-Based Testing in JavaScript/TypeScript. Annotate your functions with properties they're supposed to have, run `pabst test` to try to invalidate those claims, and watch the bugs come crawling out. Install via `npm install pabst-checker`,

Watching Terence Tao's cleanup of a Lean proof to meet Mathlib guidelines, I'm struck by the problem we're in of there being too much math. There are so many good Lean proofs coming out but essentially only one blessed outlet for them. "MathOverflow" indeed! www.youtube.com/watch?v=l3SC...

youtube.comGolfing and stylistically aligning a proof using Claude CodeIn this video, I demonstrate how an AI agent can be used to help check if a body of Lean code (intended for upstreaming to Mathlib) can be "golfed" (streamlined in presentation), and aligned with the…

Jesse Alama reposted

View full thread

Currently, we support JSDoc-style assertions such as `@total`, `@pure` to express the totality (program terminates on all arguments without exceptions) and purity (no mutation) of JS functions. But that's just the tip of the iceberg. Stay tuned for more assertions and program properties.

A longer-term aim is to complete the deep embedding of TypeScript into Lean (which will exist alongside the current shallow embedding), which will enable all sorts of features such as LSP interaction. One can even imagine an MCP server that can help harden TS programs with AI assistance.

Thales development proceeds along two arcs: (1) growing the subset of TS that we support, and (2) surfacing Lean and its program verification toolkit in TS programs, which is done via JSDoc-style comments.

As Lean becomes a load-bearing system, there's increasing pressure to makie sure that the kernel is safe.I love the idea of having an extensibile kernel and applaud the new kernel arena effort to increase transparency around that.

Claude Code helps me so much with the household chores. The dishes get washed more frequently, the apartment is appreciably tidier, the vacuum cleaner spends an above-average amount of time turned on, the trash gets taken out more quickly.

Incredible results with #lean and Gemini Deep Think by Tao in the Erdős Problem. It feels like a lot of doors in the formalization space, whether that's math or hardware/software verification, are opening quickly. mathstodon.xyz/@tao/1155914...

mathstodon.xyzTerence Tao (@tao@mathstodon.xyz)Over at the Erdos problem website, AI assistance is now becoming routine. Here is what happened recently regarding Erdos problem #367 https://www.erdosproblems.com/367 : 1. On Nov 20, Wouter van…

I was reminded of the Blueprint tool for sketching mathematical profs in Lean. I haven’t used it yet but I’m going to see if this can help me structure my work. Also curious to see how Blueprint applies in other scenarios such as software verification.

github.comGitHub - PatrickMassot/leanblueprint: plasTeX plugin to build formalization blueprints.plasTeX plugin to build formalization blueprints. Contribute to PatrickMassot/leanblueprint development by creating an account on GitHub.

Some reflections by Terence Tao on different levels, of formalization and rigor in math. This was offered as a discussion topic in the afternoon session of the Lean workshop at ITP 2026 but we didn’t really get around to it.

terrytao.wordpress.comThere’s more to mathematics than rigour and proofsThe history of every major galactic civilization tends to pass through three distinct and recognizable phases, those of Survival, Inquiry and Sophistication, otherwise known as the How, Why, and Wh…

Just had my first success using the new grind tactic in a #lean proof. I didn't know it was even going to work; I was stuck and reached for it only as a hail-mary desparation move. It worked! Mind blown.