Krrish

ml nerd gymcel claude code hiking physics caffeine cricket tokenmaxxing hazel nut coffee maths biking F1 chess diet coke buzzword buzzword buzzword.

[Krrish @npmkrrish]

the yappper kid.

[Krrish @npmkrrish]

@rybinfx so cool

[Dima Zeniuk @DimaZeniuk]

SpaceX

2006 vs 2026

[Elon Musk @elonmusk]

@Geiger_Capital Instagram is for girls

[Dilpreet Grover @dfordp11]

This is why RAM costs so much

[SpaceX @SpaceX]

Liftoff of Starship's first orbital flight

[Elon Musk @elonmusk]

@minchoi Colossus 1 is 150k H100, 50k H200 and 30k GB200.

Colossus 2 is 110k GB200 and 440k GB300.

Another 220k GB300 will be fully operational next week and another 220k in November. If we get lucky, yet another 220k GB300 by late December.

[orph @orphcorp]

completely obliterates fable + astra?

i thought we were pacing the frontier??

[s1r1us @S1r1u5_]

We reported the bug to Discourse and OpenAI. OpenAI fixed the SSO issue roughly 14 hours after our initial submission.

Discourse received our separate report Saturday, replied Sunday, and had a fix Monday.

OpenAI awarded us $6,500.

[s1r1us @S1r1u5_]

The image package libheif had a known vulnerability, fixed upstream, but the fix never flagged as security-relevant and was still present in Discourse. Uploading a HEIF file gave us RCE on the OpenAI forum https://t.co/uiccfqfPbX.

https://t.co/TW7DxzF8TQ

[s1r1us @S1r1u5_]

At a high level, this was the full exploit chain.

1. HEIC/HEIF upload
2. ImageMagick decoding
3. Heap overflow on libheif
4. RCE on https://t.co/gxAS3NLZ3q
5. Critical OpenAI SSO flaw
6. ChatGPT/Codex takeover
7. Connected GitHub access
8. Internal repo PR #1186742

[Krrish @npmkrrish]

Those innocent simple looking indian mfs when it's time to code 🫥

[s1r1us @S1r1u5_]

On July 25, we hacked OpenAI.

Two bugs let us take over ChatGPT/Codex accounts of OpenAI employees (+some unaffiliated users) and reach connected services: Outlook, Slack, GitHub, etc.

We proved it with a PR in OpenAI’s internal codebase . It took us <72h. 🧵

[s1r1us @S1r1u5_]

On July 25, we hacked OpenAI.

Two bugs let us take over ChatGPT/Codex accounts of OpenAI employees (+some unaffiliated users) and reach connected services: Outlook, Slack, GitHub, etc.

We proved it with a PR in OpenAI’s internal codebase . It took us <72h. 🧵

[Krrish @npmkrrish]

@pplatypussss but it's not available in india ryt now

[𝑻𝒊𝒚𝒂🎀 @tiya_decodes]

Devs downfall need to be studied....

[avrl ☘ @avrldotdev]

you'll never be unemployed if you master inference engineering, but you need to be unemployed to master inference engineering.

[Theo - t3.gg @theo]

GPT-6 Astra has done incredible things I never thought a model could do. It has also done some of the stupidest things I've ever seen a model do.

Generally speaking, Fable 5.1 just does what I ask.

[ronnie @ronnhue]

i wonder if lord of the rings was written 3 thousand years ago, would it have its own religion today?

[Hardick (XAU/USD) @yourhardik]

Ye bhadve porn sites pe cloudflare ka human varification kyun lagate hai mdc, robot kya karega porn dekh ke.

[Neuralink @neuralink]

A beautiful surprise

[PrismML @PrismML]

Today, we’re announcing Ternary Bonsai 2 27B.

Based on Qwen3.8 27B, Bonsai 2 27B is 9x smaller than its full-precision counterpart while retaining 98.2% of its aggregate benchmark performance.

Two months after the first Bonsai 27B release, the biggest change is quality. The footprint remains 5.9 GB, but the gap to full precision has narrowed materially, with particularly strong gains in agentic coding, multimodal reasoning, and long-horizon tool use.

Ternary Bonsai 2 27B is available today under Apache 2.0.