Sarthak

Building:

with smaller tools. Open to collab.

[Sarthak @sarthakcodes]

@steipete Is it? I was looking at ccusage data, I feel claude does a lot more re-read, but they should be priced much lower. FYI, there are 2 claude, 1 codex accounts (used to full to my best effort)

https://t.co/JaSRi8Xmzx

[Sarthak @sarthakcodes]

my claude is reading a looot more cache. how frequently do u guys /compact?

and are there any better solutions?

[Sarthak @sarthakcodes]

my claude is reading a looot more cache. how frequently do u guys /compact?

and are there any better solutions?

[Sarthak @sarthakcodes]

@pbteja1998 Yeah, hard to imagine, me paying this much

[Sarthak @sarthakcodes]

> Claude usage limits resets

Me:

[Sarthak @sarthakcodes]

@jackfriks try this cli tool
https://t.co/tuoKa8MzLR

[Sarthak @sarthakcodes]

@lennysan @LanceTMason Do you think it could have been biased?

[Sarthak @sarthakcodes]

4 pre-trained models in the gallery if you just want to play, or train your own from scratch in 5-15 min and watch the loss curve fall. Curious what you train it on.

downstream of @karpathy 's nanoGPT in spirit. cc @greggman if you find the WebGPU side interesting

[Sarthak @sarthakcodes]

watch a transformer learn English in a browser tab.

without servers or any API, just your laptop doing real math.

[Sarthak @sarthakcodes]

A few things that made this actually work: Flash Attention 2 in WGSL, weights stored packed-fp16 with fp32 accumulators (gated by a numerics check on your actual GPU), and a SharedArrayBuffer + worker setup so the UI stays responsive while training.

[Sarthak @sarthakcodes]

I have 1 user (me) for all my internal products. And this I what I see for last 30d, so a loooot of bot traffic in my opinion.
https://t.co/fJgKmZn4oK

[@levelsio @levelsio]

I have 300 million users and 111,902 customers

The SaaS services I replaced with AI are used by all those users and customers!

[Sarthak @sarthakcodes]

Then it created an issue in the repo: https://t.co/v6kNAqjyKC

[Sarthak @sarthakcodes]

There was a sudden spike in my storage around 100gb according to my calculation, asked Claude to debug. Here's what it found:

[Sarthak @sarthakcodes]

Asked why it happened:
107GB of logs were written in ~3 minutes (10:55 to 10:58 today). Multiple subagents (ae85830b839d, ac1de4986f09, a139e167925a) were all writing to the same output file simultaneously, and they were doing things like curling websites
(Scraping career pages for your resume tailor).

The bug is that Claude Code's task output system was appending every progress tick and every message from all subagents into one shared file at an absurd rate — writing ~600MB/second of JSONL logs. This is a Claude Code bug, not something
You caused.

It all happened today around 10:55 UTC (likely when you last used the resume-tailor project).

[Sarthak @sarthakcodes]

https://t.co/eFI2s07YNe

[Sarthak @sarthakcodes]

I had too many subs in my mail. And Gmail does not have a super convenient way. So I build a small tool to show me some basic analytics and help me unsub easily. It just needs reach access and doesn't store anything, feel free to give a try if you have this problem too.

[Sarthak @sarthakcodes]

When I share my vibe-coded product, do I add the "Made with AI" tag to it? 😂

[Sarthak @sarthakcodes]

https://t.co/eyCmTO8MCD

[Sarthak @sarthakcodes]

Releasing a small tool to arrange your GitHub stars into lists and tags. Because GitHub lists hardly work for me, I have ~1k stars.

You can even share your lists and possibly review other repositories. Let me know what features would you like to have in this