Vrash eu/acc (@vrashabh) profile picture

Founder and CEO https://nitter.cf/t.co/zFF2lEiM6u . Audhd. DM for startups/investing/advice. Londonmaxxing. ex carwow/Amazon/yc@ImagineK12. Trading bot

London, England
Joined April 2009
Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)Photo by Vrash eu/acc (@vrashabh)
Stoked to be on the cover & have an article published in "97 things every engineering manager should know"! Thanks @skamille . Look fwd to reading the other 96 pieces from the likes of @lara_hogan , @Lethain and @akashb . #Engineeringmanagement ♥️ Buy: amzn.to/2RASo9N
24
Vrash eu/acc retweeted
The Lean FRO and Mathlib Initiative have released a joint statement on the Lean verification of the Riemann Hypothesis. On November 20, Levent Alpöge released a proof of the Riemann Hypothesis as a Lean library of 17 million lines of code. Over the past day, the Lean and Mathlib communities worked together to verify it. The statement certifies the proof as correct and describes how it meets all necessary verification criteria. Congratulations to Levent on this achievement, and thank you to everyone in the community who contributed to the verification. Full statement: leanprover-community.github.…
471
4,922
1,926
26,367
2,792,756
Vrash eu/acc retweeted
average day in the singularity
45
625
50
6,553
128,803
Lol
Honestly that’s a long time to do the dishes
32
Lol
the commute to the anthropic office looking a little different today
41
Well, that's... One way to curate the headline
BREAKING: NVDA erects massive green candle amid reports that Jensen Huang is discussing a comeback as CEO.
1
1
57
Vrash eu/acc retweeted
Kind of insane how anytime a major news event happens, this app puts the most egregious fear-mongering content on the For You tab like we’re all going to be dead by 9pm
219
946
61
17,906
372,458
Vrash eu/acc retweeted
via Internet
33
44
12
438
53,354
Vrash eu/acc retweeted
To the Microsoft employees desperately trying to bring Outlook back online, please don’t push yourselves too hard. Your health matters more than any deadline. Take a long weekend, relax, and come back Monday feeling refreshed. You’ve worked hard enough this week, and I promise those 400 unread emails aren’t going anywhere.
739
28,456
1,604
168,118
4,295,109
Vrash eu/acc retweeted
there’s a UK-based AI data broker that no one is talking about Brighton and Hove Albion F.C.
7
7
1
346
31,451
Thanks Jim going long on alphabet
Why did Berkshire take such a big, ill-advised position in Alphabet?
33
Lol
on the worst day for the stock market all year, $DNUT is up 42% guess everyone is rotating into stress eating
41
Vrash eu/acc retweeted
give a man a fish and you feed him for a day teach a man to Claude Code and you financially ruin him with distracting side projects for the rest of his life
415
1,669
203
25,749
644,084
😂😂😂😂😂😂
chamath gonna have pronouns in bio by noon tomorrow
44
😂😂
Adderall status: 30 mg taken Modafinil status: 400 mg ingested Black coffee status: 32 oz quaffed Nicotine pouch status: 15 mg under lip I will be the first human to become Opus 5.5
43
Interesting
It looks like Astra didn't natively read Qwen's weights? As I understand it, Qwen would get stuck in a loop and not answer the question. Astra used code to perform some standard interp techniques, found a looping direction, and ablated it from the weights.
28
Holy shit
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math. rikiyat.github.io/blog/weight-sp… github.com/RikiyaT/lm_und… huggingface.co/Qwen/Qwen3.5-0… huggingface.co/RikiyaT/Qwen3.…
29
Black Friday market tmrw? Just wild
2
33
Umm what? So.. eli5, For 25 years, a core assumption in computing has been that some problems get exponentially harder as they grow. An OpenAI model claims a shortcut. If verified, it won't make them easy, but could change our understanding of computation. But maybe this is where AI doing original math? Not ethereum so don't go short it (yet lol)
An internal model at OpenAI has produced a claimed refutation of ETH! Russell Impagliazzo, Mohan Paturi, and I have been asked to help organize an open and independent examination of this claim. 1/4
38