mohsen1

7 ms·

91.
by mohsen1 5mo ago
in tsz ( https://tsz.dev ) I am Codex-Maxxing with this: Give each Codex an AgentName and ask them to mark their PR/issue/comments with those. Have one or two "managers" that manage PRs and overall project dire
92.
by mohsen1 5mo ago
This is much needed! Compared to Codex CLI, Claude Code is insanely slow. $ time claude --version 2.1.143 (Claude Code) ________________________________________________________ Executed in 4.39 secs fish
93.
by mohsen1 5mo ago
In the Sound Mode page I mentioned that I have some ideas around this. What if we could take .d.ts and project a sound version out of it? How feasible that can be is unclear but the biggest unsoundness criminal is `any` which is relatively
94.
by mohsen1 5mo ago
Thank you for the feedback. I think I should rename the "Sound Mode" to something more honest. "Strict" is already taken so I should think of a better name. Making tsz compatible with tsc and making it sound is mathemati
95.
by mohsen1 5mo ago
tsz for me is an experiment to see how can this kind of work be done better. With a slight difference that tsz is not a direct port and it's a different architecture. I'm also not claiming to have answers but I've learned a
96.
by mohsen1 5mo ago
I was a little shocked that they could get it fully working in a week to be honest. My side project is a very similar ambition ( https://tsz.dev ) but I am in no way claiming success. i keep adding more and more tests to ensure th
97.
by mohsen1 5mo ago
in tsz[0] 100% of tests pass yet I have a ton of bugs. I don't think any software out there is fully tested really. I'm experimenting this this idea as well. So far learned a ton. I'm convinced the future of writing code is h
98.
by mohsen1 5mo ago
deleted
99.
Sound Mode: Can TypeScript Type Checking Be Stricter?(tsz.dev)
6 points by mohsen1 5mo ago|2 comments
100.
by mohsen1 5mo ago
Because I don't know those languages. I'm still reading the code LLM writes
101.
by mohsen1 5mo ago
https://github.com/mohsen1/tsz
102.
by mohsen1 5mo ago
Yes it does. Usually after hours of working and not getting results
103.
by mohsen1 5mo ago
I don't know about Mythos but in recent weeks I've noticed Opus is constantly failing to fix things in tsz[0] vs GPT 5.5 can easily churn out fixes that are solid and pass tests. I've stopped paying for Claude for now and all
104.
by mohsen1 5mo ago
You can see LSP working in the playground. It is still WIP but pretty fast. I'll add LSP benchmarks later. My design is highly biased towards fast incremental checks
105.
by mohsen1 5mo ago
I'm taking a different path than Enzo. Fully tsc compatible and more sound on top
106.
by mohsen1 5mo ago
Working on tsz ( https://tsz.dev ) a performance-first TypeScript checker written in Rust. Started 5 months ago and it's been mostly AI-written code. 99.8% tsc conformance test pass rate today. Single file benchmarks are 3–5x
107.
by mohsen1 5mo ago
That talk is very interesting. Thanks for sharing. I'll watch it later. Now that most of the implementation is nearly completed, I'm building a lot of instrumentations to have better visibility into those things to have concrete a
108.
by mohsen1 5mo ago
Full soundness in JavaScript is proven to be impossible. You can write code that no compiler can really follow the execution path. There has been attempts in the past ( https://hegel.js.org ) but I think TypeScript has been succes
109.
by mohsen1 5mo ago
Very cool! If you have extra tokens laying around ask the agent try to break things and open GitHub issues. This is what I do for tsz and beyond conformance test I can see it finding very good bugs.
110.
by mohsen1 5mo ago
From the languages that I know, Rust is the only language that I can look at a multi-threaded code and understand it. This stuff being checked by the compiler is a huge advantage
111.
by mohsen1 5mo ago
OpenAI gave me that 10x boost and used it all already for this week. I'm guessing the last 50 tests is only doable by GPT 5.5 xhigh
112.
by mohsen1 5mo ago
Thank you! It's much faster in single file benchmarks (3 to 5x) https://tsz.dev/benchmarks/micro I have optimizations planned for large projects that I'm still flushing out.
113.
by mohsen1 5mo ago
Yes indeed. More than 1 million lines of code (including tests) is jumping lots of hoops but with LLMs it's not as painful so you can just ask it to do the hard things. Example of a Claude Code session after 2 hours of "Crunching&
114.
by mohsen1 5mo ago
Very impressive that they could do this so quickly because I have been on a similar project (porting TypeScript to Rust) for 5 months. But I guess I don't have access to Mythos and unlimited tokens. I'm also close to 100% pass rat
115.
by mohsen1 5mo ago
I've been using LiveKit which is also WebRTC based and it is super annoying when speed slows down or speeds up at times when connection is not robust. We were using OpenAI's websocket based RealTime audio which was way too slow. S
116.
by mohsen1 5mo ago
More or less. The tsz solver is pure enough (it doesn't know about the AST) that it might be possible to formally validate it. But in my case I am lucky with tsc baseline. Anything that produces different output than tsc is a bug
117.
by mohsen1 5mo ago
I think the same issue from a couple of days ago is resurfaced because I can't see Pull Requests that were opened just minutes ago and merged PRs are still showing in the list: https://github.com/mohsen1/tsz/p
118.
by mohsen1 5mo ago
Code is not the goal. What code does is.
119.
by mohsen1 5mo ago
For me, unless there is a concrete way of proving work is correct you can't rely on AI coding. tsz has super strict tests around correctness, performance and architectural boundaries
120.
by mohsen1 5mo ago
Multiple computers and each multiple Claude Code or Codex sessions. It had lots of ups and downs. Now I have a good enough test harness that makes it easier to iterate faster

More ›