Ankit

Locked in. Math + AI + Algorithms Research.
Mom's gajjar halwa & aloo paratha enjoyer.
Jai Jagannath!

[Destroy the idea of Matt Nagy @GigaflyStalin]

0.4 GPA activities

[˚꩜｡ @yappingwmogu143]

not to be a fucking nerd or anything but writing academic papers is actually super fun when you have a genuine interest in the topic

[Roberto Blake 🇺🇸🇵🇦 Creative Entrepreneur @robertoblake]

I want more data centers.
I want more nuclear power plants.
I want more water desalination plants.
I want more high speed transport.
I want more automated warehouses.
I want more 3D printing.

I want an abundant, advanced, high trust society… no longer stifled by doomerism, neuroticism and cowardice.

[@dingchilling 🫪 @dingchilling]

gemini search is still undefeated

[Ankit @sudoankit]

@ShivAroor @VishnuNDTV 😂🤣🤣😂😂 bruh

[Hemant Mohapatra @MohapatraHemant]

The curse of discipline is that every day looks the same. The curse of indiscipline is that every year does.

[Ankit @sudoankit]

@shitpost_2077 The song is “Animals of the Farm” by Saavane is 6’4

[Martin Bauer @martinmbauer]

Innumeracy. Being proudly “bad at math” is socially acceptable in a way being proudly unable to read would never be

[Jacob Titus @jacob__titus]

Forbes on manufacturing.

[Ankit @sudoankit]

@dpradhanbjp @RahulGandhi @INCIndia lmaaoooo AI reply

[Ankit @sudoankit]

👌🏽

[jo johnson @josbjohnson]

GOD WOULD NOT PLANT A DREAM IN A MAN INCAPABLE OF REACHING IT.

[Ankit @sudoankit]

@d_goryainov Incredible work, big fan.

[David @DavidSHolz]

you gotta be confident enough to act and open enough to learn

[Tai Lopez @tailopez]

Seneca was right when he said to be happy you must eliminate 2 things: the fear of a bad future and the memory of a bad past.

[Ankit @sudoankit]

@Dhruvam987 🤡 levels of engagement farming

[🫧 @lunecrumbss]

lanterns tree 🏮

[Sahil Bloom @SahilBloom]

I'm convinced that the single greatest challenge for any ambitious person is eliminating the guilt associated with free time and rest.

[Aleksa Gordić (水平问题) @gordic_aleksa]

new in-depth blog post time: Inside the Transformer: The Life of a Token

a deep dive into a modern dense transformer, i cover YaRN (why does pairwise coordinate rotation induce positional information?), hybrid attention (getting to 160k context length), soft capping, QK normalization, etc. as the token flows through the transformer

bonus transformer math: FLOPs/token formula (and when is 6N formula broken), cluster sizing (how big of a cluster do you need given the model/data size and experiment throughput of interest), and more