Afroz Mohiuddin

@_arohan_ All the very very best!

Ok I did leave anthropic, a few weeks ago, it was one of the best places to work for a researcher. Jerry Tworek nerdsniped me into starting this with him and others The pretraining team at ant is one of the well functioning research team in the industry, and anthropic has great culture - I miss the fun times and claude! Thank you for everything!

@AndrewDai @nmasc_ Well reasoned, much respect!🫡

@nmasc_ Thanks Natasha! We really value the team both current and future. It felt unfair to new employees to force them to accept a high valuation.

I remember writing a column last year about VC investors no longer backing model startups because of the heavy costs and competition from big labs. That perspective has significantly changed as reasoning and specialization matters more and more. Our latest scoop on @technology digs into yet another effort Latest w/ @byJuliaLove: bloomberg.com/news/articles/…

Many congratulations @nitishkulki and @alankarjain91 -- This is awesome and can't wait to see how this is used in the wild!!

“The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function” — F Scott Fitzgerald

@JingyuanLiu123 @Jianlin_S @clu_cheng Congratulations on the move @JingyuanLiu123 ! You’ll do great and happy for you!

Some updates: I've always been bullish on TML, and I actually joined TML this Monday Looking back, I am feeling so lucky that I have the privilege to work closely with the best optimization experts on the Muon optimizer ( @Jianlin_S from Kimi and @clu_cheng from Meta). Now I am so excited to be able to work with @jxbz and build new cool things! (On the other hand, there have always been some bad rumors about Meta TBD's potential failure. That's not true! From my personal experiences, it really has the best talents in the field, and I really enjoyed learning from the lab. The avocado model will for sure be great!)

@agihippo Useful!

What's the highest utility per per dollar / ROI things you bought /subbed in the recent years? here's mine (i don't spend much money in general but here's my tier list anyway): S-tier [outsized value / insane] - fitness tracker / pixel watch. 300 dollars and tracks random metrics but the key thing is that it spurs you on to be more health conscious (sleep and exercise) - totally worth it. superb value for money. can't find anything more ROI efficient than this. - good quality snacks that are high in protein (sometimes pricey for snacks but good health is always worth it). For example, high protein ice cream is like 2.5x more expensive, tastes slightly worse but has decent macros. A-tier [pretty good / amazing] - car. A simple tesla in Singapore costs 200K but i found it to be so high ROI somehow because I can travel to play sports easily or go anywhere easily which improves my health & time significantly (two most important metrics!). MRT / grab is ok but somehow even though they are pareto-efficient it's still different driving. A luxury car might be C tier though but I think an affordable EV is super high ROI. - Michelin star fine dining with family. (~800 SGD for 2 people for an unforgettable meal) not bad for good experiences. occasionally though like once a month, not everyday. - X premium. Mostly any other subscription like YT premium, netflix etc. The cost to watch ads is just higher than any cost you pay for subscriptions. - home gym (we paid for a home barbel/dumbbell set years ago for $3K and did more than a thousand sessions combined on that thing). needs some floor space though. - meal prep service. $10 per meal and they ship good quality meals with reasonable stats to you. slightly troublesome but probably good value compared to eating on grab food every single meal/day. B+ / B / B- tier "okay-ish good to okay-ish maybe." - business class tickets for flights >6 hours [B+ tier] necessary for sanity even on your own dime. - top end sports equipment [B+ tier] (one thing i learned in sports like badminton is just buy the expensive rackets anyways instead of finding one like tens of dollars slightly cheaper. - macro tracking app [B+ tier]. (sure, you can track in a spreadsheet too but this spurs you on to be more health conscious). - grand piano (~30k). I don't use it that much even though I used to be a classical pianist but my wife does. but b tier for the vibe enhancements to the house. [B tier] - supplements like vitamins/magnesium/ashwaganda. Makes you feel like you're healthmaxxing but I honestly don't feel any difference even after a few months. They should be based on science but I didn't see any effects except placebo maybe. [B- tier] - personal homepage. I use squarespace and pay like $100 per year or so? I swear I could have used github pages and it would do the same thing but anyways. Not the greatest thing but not the worse [B- tier] D-tier (mostly cancelled or never used) - Linkedin premium (I used to have that on a corp card) but I wouldn't pay for it myself. Useless tier. - Tesla add-ons in Singapore (i don't need to pay extra money i can park myself lol). There's no FSD in Singapore either. - Meditation app (i used one but it was good but time consuming so i deleted and unsubbed) - Ipad or tablet. Between a computer and a phone, what is the point of a tablet?? Things not sure and still wondering - smart ring (i was always tempted to buy one but i have a smart watch and felt it's a waste of money). Am I wrong? - very fancy gym memberships (maybe like 200 SGD per month?) i never got down to sign up because google has a gym and i have one at home too). But are these cold baths, saunas and good lifting vibes worth it? idk. - sleep optimization things like temperature control blankets/sheets etc. (are these worth it?) - probably missed more here. - personal assistant (@swyx suggested this at one point but im still wondering if it's more time cost than doing stuff yourself). i'm looking for things that are high ROI / good pareto frontier to buy and do some weekend retail therapy but have no clue...lmk if any suggestions.

@agarwl_ @urvashi__s Many congratulations to both of you!

The highlight of my year happened right before it ended. Happy new year!

This one Munger quote is so counterintuitive, yet empirically so accurate.

“At my age, you make new friends, or you don’t have any friends,” Charlie Munger during in his last, eventful years. wsj.com/finance/invest…

Thanksgiving-week treat: an epic conversation on Frontier AI with @lukaszkaiser -co-author of “Attention Is All You Need” (Transformers) and leading research scientist at @OpenAI working on GPT-5.1-era reasoning models. 00:00 – Cold open and intro 01:29 – “AI slowdown” vs a wild week of new frontier models 08:03 – Low-hanging fruit, infra, RL training and better data 11:39 – What is a reasoning model, in plain language 17:02 – Chain-of-thought and training the thinking process with RL 21:39 – Łukasz’s path: from logic and France to Google and Kurzweil 24:20 – Inside the Transformer story and what “attention” really means 28:42 – From Google Brain to OpenAI: culture, scale and GPUs 32:49 – What’s next for pre-training, GPUs and distillation 37:29 – Can we still understand these models? Circuits, sparsity and black boxes 39:42 – GPT-4 → GPT-5 → GPT-5.1: what actually changed 42:40 – Post-training, safety and teaching GPT-5.1 different tones 46:16 – How long should GPT-5.1 think? Reasoning tokens and jagged abilities 47:43 – The five-year-old’s dot puzzle that still breaks frontier models 52:22 – Generalization, child-like learning and whether reasoning is enough 53:48 – Beyond Transformers: ARC, LeCun’s ideas and multimodal bottlenecks 56:10 – GPT-5.1 Codex Max, long-running agents and compaction 1:00:06 – Will foundation models eat most apps? The translation analogy and trust 1:02:34 – What still needs to be solved, and where AI might go next

Shortcut [noun]: The longest path between two points. We love them around here! 😂

@prajdabre More about Warren Buffett disclosing his stake I’d guess!

Oh boy!

In 1890, 23 years old without any formal degree, she was working as a governess in her father's friend's house in Warsaw to help support her sister who was struggling to get an education in Paris. She fell in love with the elder son of the family who refused her as she had no prospects. Heartbroken and weary she wrote to her sister Bronia that she was stupid, would remain stupid and nothing would ever come of her life. Thirteen years later she won a Nobel Prize in Physics along with her husband Pierre, and won another one for Chemistry eight years later. On Marie Curie's birthday, giving up shouldn't ever be an option.

Sometimes examples like these drive home the Sherlock Holmes quote — “You see, but you do not observe. The distinction is clear.” (3/3)

If a bacteria doubles every minute, and in 60 minute fills up the container — Exactly when does a bacterium first realize this fact? Even a mere 5 seconds before, the container is only 3% full, ie 97% empty! (2/3)

“Our inability to understand the exponential function is our biggest weakness” — Prof Albert Bartlett youtu.be/1bvwOrGn1Zs?si… The example given is striking in its simplicity! (1/3)

Amazing work and gains by a solid set of folks!

@_arohan_ Heavy lifting word …

@afrozenator “Early” is working hard. Very much agree.

I think outside early Google employees it’s not appreciated as much how academic the early Google culture was, open internally, ambitious and excited about nerd sniping each other — It was the best place one could land up in for my generation even … Truly a magical place.

Congratulations @achowdhery and team!