🐍 Tal Weiss (@majortal) header
🐍 Tal Weiss (@majortal) profile picture

Founder https://teapawt.net/t.co/qmgZ1kiSY6 Built Evature β†’ acquired by Booking. Ran ML there. Building AI for agents // old people. Mostly tech. Shipping: https://teapawt.net/t.co/lN3g94T9q5 https://teapawt.net/t.co/seKA5V1QZj

Amsterdam, The Netherlands
Joined February 2008
Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)Photo by 🐍 Tal Weiss (@majortal)
What will we call the Riemann hypothesis now?
1
1
2
134
🐍 Tal Weiss retweeted
Did someone check to make sure the Riemann hypothesis proof wasn't in the training set?
39
34
7
1,099
35,653
🐍 Tal Weiss retweeted
Oh boy … PHP sent clear text data when you used uppercased S in httpS:// in a protocol handler πŸ™ˆπŸ™ˆ file_get_contents('HTTPS://…') skipped TLS: tokens, POST bodies and FTPS logins went out unencrypted! Fixed today in 8.2.36, 8.3.37, 8.4.28, 8.5.13 & 8.6.0.
192
1,229
500
6,483
1,166,264
My wife's grandmother Hinda was a dressmaker in GrΓ³jec, Poland. In January 1933 she got a passport for one journey: to Palestine. She landed at Jaffa on 6 March. It says she was born in 1919, the last digit written over another. Her 1939 marriage certificate says she was 26.
1
6
169
Terraforming has begun
this might be the craziest thing humans have ever tried lol a startup called Real Ice is literally trying to refreeze the Arctic and on a small patch in northern Canada, it's actually working! > in winter, they drill through the ice and pump seawater up on top > the freezing air turns it into a fresh layer, so the ice goes into summer thicker > their test patch came out about 50 cm thicker than the ice around it this is huge because right now the Arctic is on track to lose almost all its summer ice as soon as the 2030s and that ice helps keep the whole planet cool it's like wearing a white t-shirt vs a black sweater on a hot day right now the ice is the white t-shirt, bouncing the sun's heat back into space once it melts, the dark ocean underneath becomes the black sweater, soaking all that heat up and that extra heat speeds up global warming, which means rising seas, stronger hurricanes and deadlier heatwaves all around the planet the company says thickening an area twice the size of California would cost about $5-6B a year we're officially in the business of refreezing the planet
1
1
150
Kinvestigator: The 1 Dec 1945 civil marriage of <REDACTED> is in book C24/15 at the LitomΔ›Ε™ice archive, which is not online; the papers filed with it may give the name of his first wife. Two letters to the archive have no answer yet, and it answers private requests in undigitized books only gradually, so a genealogist reading the book in person may be faster. Either letter asks for the entry, the documents filed with it, and a quote before any work. Write to a genealogist near LitomΔ›Ε™ice to read the 1945 marriage entry?
39
The safety gate didn't work
2
53
I'm still surprised my engineering manager's nose works with non biological developer smells. But are you fixing the ROOT cause? No... Will we be surprised the next time this happens? Yes... Did you consult the style guide? No... If you were me, would you have been happy with this design? No... Can we make it faster and cheaper? Yes...
1
1
1
94
Somewhere along the way, Claude appointed itself the unpaid compliance officer of every website on the internet
2
1
2
130
back to this shit?
2
89
Updated my Github profile :) Member for 17 years, 8 months
1
166
Is the end game of recursive self improvement when a model edits its own weights?
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.…
1
2
106
🐍 Tal Weiss retweeted
In case you missed it, integer factorization has been proven subexponential (deterministically, worst-case bound)! The hard work comes from Claude's version of the ETH refutation but it seems a little more had to be done to get this result for integer factorization.
22
102
12
941
58,273
🐍 Tal Weiss retweeted
Surprise 1: Anthropic acquired a one-person biotech . . . Surprise 2: that person is me Surprise 3: I’m here to tell you all about it Yes, this is what you think it is; after 6 yrs of increasingly loose anonymity: the Professor Oak reveal post …and yes, it has gotten so boring around here that I’m literally doxxing myself to spice up the timeline ------- Ok, allow me to introduce myself. My name is Cody Tranbarger. I mostly tweet, but occasionally find time to dabble in drug dev & starting/building biotechs. My past lives span the VC, crossover, startup continuum; Prof Oak has been my way of keeping a digital foot in those worlds, despite having my physical feet planted elsewhere. The past 18 mo, β€˜elsewhere’ was Mithril Medicine, a co. I founded to develop bsAbs for LAMA2-CMD & other ECM-related genetic diseases. Mithril was born out of hands-on building with @MikeNGladstone, @JasonPaulRhodes, @LifeSciVC, & team @atlasventure. Somehow, I made enough half-decent calls along the way (even a broken clock is right twice a day, eh?) to convince them Mithril/I warranted a small seed. While I can hardly rationalize such an outrageously cavalier bet, I’m grateful for it, & I'd partner with Atlas again in a heartbeat. In typical AV fashion, Mithril’s primordial form was virtual, decentralized, & ultra-lean. This stage tends to be short-lived, but for Mithril, it worked surprisingly well, for surprisingly long, thanks in part to the rapid maturation of the AI tools I used liberally. Eventually, Mithril became a bit of an experiment itself - to find the upper bound of how far it could scale as a single operator + a bunch of Claude (+ a stellar bench of part-time consultants). In parallel, I started spending time with @nc_frey, @an1lam, & @CoefficientBio team; and promptly experienced, for the 1st time, AI at the frontier. Soon thereafter, Mithril stopped looking like an experiment, & started looking like a preview of an entirely new operating model. From that spark came β€œPipeline-in-a-Person.” The idea immediately resonated when we shared it here back in April; and it was central to the shared vision that led to Anthropic’s acquisition of Mithril, & my joining the team. Why join Anthropic? Three reasons. (1) For Mithril, the programs continue, with more resources & infrastructure behind them, and as much Claude as I can put to work. (2) For me, the chance to start several more programs, & find out how many one person can actually run in parallel - an almost unprecedented luxury for a curious drug hunter. (3) For our mission, a path to institutionalize & scale this new way of working, then put it in the hands of all of biopharma, all over the world - which, candidly, I find the most energizing. Will AI x Bio cure diseases? Unequivocally yes, imo… but not by the route that gets the most airtime. Rather than timeline acceleration & compression, I see orchestration & parallelization as today's low-hanging fruit. Most drug hunters see a handful of go/no-go calls in a career, & spend entirely too much of the rest trading emails (my inbox can attest). Give them the bandwidth to run several programs at once, and the time to focus on scientific judgment, and I’m willing to bet they’ll make better calls. And at scale, better calls translate to more drugs, higher hit rates, broader patient impact, & likely, a disease or 10 or 100 cured. I couldn't be more excited to build toward that future @AnthropicAI. Finally, a heartfelt thank you to BioTwitter/X. The end of anonymity is bittersweet, but if anyone’s earned a proper introduction, you all have. It’s difficult to overstate how much this app has done for me: a decade of education, plenty of friends, many job opportunities, several mentors, a few investors, even a co-founder (fmr). It was the generosity of this community that inspired me to start posting 6 years ago, after just as many years lurking. I hope I’ve been able to pay it forward, in my own small way. But, to be clear, Prof Oak isn’t going anywhere. Same beat, same themes, and now, a front-row seat to drug development at the frontier. Stay tuned for dispatches from the trenches; and remember... Gotta Cure 'Em All
533
1,067
388
14,426
1,770,873
I'm tired. New puppy. Tesla FSD is a literal life saver.
1
5
106
Openspec was great. I liked it a lot. But the models became smarter and O.S. became a crutch, slowing it down, dumbing it down. I tried to take the best ideas and simplify them. Frontier AI deserves frontier specs. kyspec.com
OpenSpec is about to get 10x better. We've spent the past year learning from developers and teams using OpenSpec every day. Now we're bringing those lessons into our own custom-built agent harness. Our goal is simple: help developers stay involved in the decisions that matter, better understand the work being done, and ultimately have greater confidence in what agents produce. A completely new way to experience OpenSpec. We'll be releasing an early preview next week. Stay tuned!
85
I asked Codex to do some math in spot instances so it'll be cheap. Apparently it emailed AWS for a quota increase for spot instances in ALL regions. Yada yada - it wasn't cheap.
3
2
296
🐍 Tal Weiss retweeted
Replying to @cohenrap @majortal
I like it! Very aligned to how I want to use specs.
1
1
1
57