Anatoly Karp
@akarp
Followers
3K
Following
218K
Media
390
Statuses
7K
MySQL Infra at Meta. Method coder. Discreet mathematician. Rust enjoyer. Machine learning, physics, economics.
Sunnyvale, CA
Joined May 2008
If any of my mutuals are in Davos now and want to meet up let me know. [I'm not in Davos, just wondering if you want to meet up]
0
0
8
that's it, impeach now
BREAKING: Trump says that if Macron is 'hostile' to Board of Peace invitation, he will impose 200% tariff on French wines and champagnes.
0
1
6
I can't bring myself to dig into Ralph, it just seems so brain-dead. Should I?
4
0
4
Did you know that Levi-Civita was actually one guy? What an absolute unit.
3
3
123
My post blew up in finite time so I'm adding another nice exposition (in French)
1
0
9
Terry Tao understood Perelman's proof of the Poincare conjectore so you don't have to
12
84
605
Yoneda is just a glorified Cayley's theorem, what's the problem?
'I've decided that the Yoneda lemma is the hardest trivial thing in mathematics.' -- @sigfpe
2
1
13
Epistemic status: blissed out
John C. Baez: Coxeter and Dynkin Diagrams https://t.co/j8sKC7wCoL
https://t.co/uyMsk3HRaD
0
1
2
Are you scared? Don't be. The Weierstrass function doesn't actually exist, silly
This is the Weierstrauss function, a “monster” that is continuous everywhere but differentiable nowhere. First published in 1872, it prompted scrutiny of assumptions in calculus. Similar functions are used to study random motion and analyze risk. https://t.co/tMXyKuJIrC
0
0
6
5 recommended Economics books, 2025 1. Truman Bewley, Price Setting This book is a bookend to Bewley's 1999 Why Wages Don't Fall During a Recession. Together, these books contain invaluable interview-based information on price determination. Remarkable work from one of the
12
374
2K
I’m terrible at picking presents but apparently people who knew me as a kid had me figured out pretty good
0
0
10
so wait a minute, that guy torched the reputation of Harvard, MIT and DeepMind in two days and then just disappeared, never to be seen again?
5
0
27
"not a mathematician by trade" huh. I think we are done here
@ElliotGlazer @NunoSempere I'm not expecting anything. Lean is a tool. It just happens I'm a computer scientist and not a mathematician by trade, so I find iterating on Lean and LLMing that into a paper later to be convenient (when it works). I appreciate Lean is hard. See my post from ~15 mins ago
0
0
4
Me after sending my vibe proof of N vs NP to Annals of Mathematics
0
1
3
Protocol: the paragraphs written by GPT5 MUST start with [Trust me bro]
This is wild! Johannes Schmitt used GPT5 to solve his own open problem on intersection numbers on moduli spaces of curves (the proof turns out to be unexpectedly simple, "low hanging fruit"). He wrote up the paper, being careful to point out which *entire paragraphs* were written
0
0
4