asaba bryan @basaba07

Move37 Joined January 2011

Tweets

9K
Followers

808
Following

854
Likes

809

Eugene Yan @eugeneyan

2 weeks ago

after leading a few projects, i've found that once you've set up the evals + experiment harness and make it easy to tweak config and prompts with 1-click run + eval, teams enjoy running experiments and hill climbing those numbers, and progress comes quickly. but setting up that…

26 52 704 65K 579

Andrej Karpathy @karpathy

a month ago

2024: everyone releasing their own Chat 2025: everyone releasing their own Code

481 595 8K 735K 652

K Srinivas Rao @sriniously

a month ago

Pattern recognition is the primary architecture of intelligence itself. When you can see the recurring structures in market cycles, human behavior, code bugs, relationship dynamics, or scientific phenomena, you're essentially running the same neural machinery that allowed our…

68 1K 5K 249K 3K

Andrej Karpathy @karpathy

3 months ago

Some of the links: - My slides as keynote: drive.google.com/file/d/1a0h1mk… - Software 2.0 blog post from 2017 karpathy.medium.com/software-2-0-a… - How LLMs flip the script on technology diffusion karpathy.bearblog.dev/power-to-the-p… - Vibe coding MenuGen (retrospective) karpathy.bearblog.dev/vibe-coding-me…

52 232 2K 261K 2K

Ethan Mollick @emollick

4 months ago

Huh. Looks like Plato was right. A new paper shows all language models converge on the same "universal geometry" of meaning. Researchers can translate between ANY model's embeddings without seeing the original text. Implications for philosophy and vector databases alike.

405 2K 13K 1.7M 11K

Download Image

ollama @ollama

4 months ago

Multimodal model support is here in 0.7! Ollama now supports multimodal models via its new engine. Cool vision models to try👇 - Llama 4 Scout & Maverick - Gemma 3 - Qwen 2.5 VL - Mistral Small 3.1 and more 😍 Blog post 🧵👇

68 280 2K 105K 470

Download Image

Gautam Kedia @thegautam

4 months ago

TL;DR: We built a transformer-based payments foundation model. It works. For years, Stripe has been using machine learning models trained on discrete features (BIN, zip, payment method, etc.) to improve our products for users. And these feature-by-feature efforts have worked…

180 557 5K 1.3M 4K

Deedy @deedydas

7 months ago

Everyone should be using this website to understand the inside of an LLM. I'm surprised more people don't know about it. Benjamin Bycroft made this beautiful interactive visualization to show exactly how the inner workings of each of the weights of an LLM work. Here's a link:

70 928 7K 710K 14K

Download Image

Naval @naval

8 months ago

Build it (well) and they will (eventually) come.

368 1K 12K 701K 2K

Andrej Karpathy @karpathy

8 months ago

Now everyone can be a super popular live streamer influencer (to an AI audience 😂) amazing

Justine Moore @venturetwins

8 months ago

Now everyone can be a super popular live streamer influencer (to an AI audience 😂) amazing

138 99 2K 750K 938

Download Video

139 132 2K 246K 432

Z Fellows @zfellows

9 months ago

"A founder is not a job, it's a role, an attitude. And it's something that can happen again and again and again, and in fact it has to happen again and again and again, otherwise we would not move forward." — Jack Dorsey, co-founder of Twitter

1 39 258 8K 64

Download Image

François Chollet @fchollet

11 months ago

A good heuristic about whether something can in principle be solved with deep learning: can it be done by an expert human 100% intuitively, without having to consciously think about it? If yes, then pure pattern recognition is an appropriate approach to the problem, and given…

1 152 1K 153K 509

Philipp Schmid @_philschmid

11 months ago

This came unexpected! @OpenAI released Swarm, a lightweight library for building multi-agent systems. Swarm provides a stateless abstraction to manage interactions and handoffs between multiple agents and does not use the Assistants API. 🤔 How it works: 1️⃣ Define Agents, each…

62 336 2K 259K 3K

Download Image

The Nobel Prize @NobelPrize

11 months ago

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”

1K 13K 33K 12.6M 3K

Download Image

Ilya Sutskever @ilyasut

11 months ago

Congratulations to @geoffreyhinton for winning the Nobel Prize in physics!!

209 619 12K 829K 360

Andrew Ng @AndrewYNg

a year ago

After a recent price reduction by OpenAI, GPT-4o tokens now cost $4 per million tokens (using a blended rate that assumes 80% input and 20% output tokens). GPT-4 cost $36 per million tokens at its initial release in March 2023. This price reduction over 17 months corresponds to…

115 630 4K 741K 2K

François Chollet @fchollet

12 months ago

My general intuition about human intelligence is that spending more time on a problem will yield roughly linear improvement as a function of solution complexity (e.g. program length), within some complexity bound. That's what I see in software engineering, in particular. With 2x…

28 47 582 57K 205

François Chollet @fchollet

12 months ago

In software, naming things matters, because names are the first layer of documentation that conveys the mental model behind the code.

64 177 2K 90K 208

François Chollet @fchollet

12 months ago

The real impact of technology often lies not in what the tech itself can do, but in how it changes human behavior and interactions over the long run. For better or worse...

27 75 415 31K 61

Andrej Karpathy @karpathy

a year ago

Programming is changing so fast... I'm trying VS Code Cursor + Sonnet 3.5 instead of GitHub Copilot again and I think it's now a net win. Just empirically, over the last few days most of my "programming" is now writing English (prompting and then reviewing and editing the…