Go · Rust · AI 💻 Payments infra by day, next-gen AI by night 🤖 Teaching machines to think cheaper and fail less 🧠 Building in silence ⚡
go.mod
Joined June 2014
- Tweets10.8K
- Following2.2K
- Followers1.5K
- Likes11.5K
Pinned Tweet
When wisdom enters your heart
And knowledge becomes pleasant to your soul,
Thinking ability will keep watch over you,
And discernment will safeguard you,
To save you from the bad course,
From the man speaking perverse things,
From those leaving the upright paths.
Most AI assumes unlimited compute. Most of the world doesn't have it.
I'm building for the other side: AI that's fast, reliable and runs on modest hardware.
Cheap means nothing if it quietly gets worse. ⚡
Kay OBA GO retweeted
holy f*ck, Karpathy put an entire AI degree on GitHub for free
7 repos, 306,741 stars, and in the right order they take you from backprop to agents that run research overnight
each repo is the next layer of the one before
1. micrograd -> how a neural net actually learns
backprop in about 100 lines of Python
↳ github.com/karpathy/microgra…
2. nn-zero-to-hero -> the video course
his lecture notebooks, from one neuron to GPT
↳ github.com/karpathy/nn-zero-…
3. minbpe -> how text turns into tokens
the tokenizer every LLM starts with, in clean minimal code
↳ github.com/karpathy/minbpe
4. nanoGPT -> your own GPT
the simplest repo to train and fine-tune one
↳ github.com/karpathy/nanoGPT
5. llm.c -> the same training with nothing hidden
raw C and CUDA, no PyTorch underneath
↳ github.com/karpathy/llm.c
6. nanochat -> a full chat model end to end
GPT-2 level training cost ~$43,000 in 2019
here it's about $48
↳ github.com/karpathy/nanochat
7. autoresearch -> agents that do the research
an agent edits the training code, trains for 5 minutes, keeps what improved and repeats all night
↳ github.com/karpathy/autorese…
train one GPT from scratch and you become the person who gets handed the AI project, along with the raise that comes with it
the classroom half of the same degree, what Stanford actually teaches about AI, is below ↓
Kay OBA GO retweeted
I spent 4 months collecting this so you don’t have to waste the same time
Most people will save this and never open a single link.
This is the clean path from zero to being able to build and interview for real AI roles
01 Linear algebra and calculus
mml-book.github.io/
02 Machine learning foundations
youtube.com/playlist?list=PL…
03 Deep learning from scratch
youtube.com/playlist?list=PL…
04 Neural networks and backprop
youtube.com/watch?v=VMj-3S1t…
05 Computer vision
youtube.com/playlist?list=PL…
06 Natural language processing
youtube.com/playlist?list=PL…
07 Transformers and LLMs
cs336.stanford.edu/
08 Reinforcement learning
rail.eecs.berkeley.edu/deepr…
09 Building agents
huggingface.co/learn/agents-…
10 The full path. Free
You either walk this now or keep watching random playlists for another year
Save this ⭣
Kay OBA GO retweeted
My dear friends, I’d like to show you my latest toy project
Introducing Handwired: turn your regular JavaScript code into a neural net!
- no training
- exact outputs vs the JS program
- fully differentiable
All runnable in a browser, without GPU
Building a tiny autograd engine in Go, inspired by Karpathy’s micrograd.
Go has no operator overloading, so every operation is explicit: a.Mul(b).Add(c). Turns out that’s a feature. You can’t hand-wave the chain rule when you have to write every backward pass yourself.
Kay OBA GO retweeted
🚨 Fireboy DML on the minimal collaborations on "ZUKO!"
— "It just didn't work out, I swear to God. And I just felt like that's how it's meant to be for this particular project. I genuinely wanted it to be very collaborative but it just didn't work out. I don't force vibes"
Kay OBA GO retweeted
Just played this album for my boss
Bro said Fireboy’s music is not for illiterates 😭
😂😂
In my memory, that’s where you are, yeah
You’re my medicine, my ritual, yeah
And I will never be the same without you, my baby o
And I feel sick when I'm away from you, baby o
@fireboydml thank you for this line mehnnnnn
Kay OBA GO retweeted
Calculus is a powerful tool for understanding change, motion, and growth. And it's good to know as a dev.
In this new College Calculus course, Ed teaches you the key concepts through the lens of Python.
You'll learn about limits, derivative rules, slope interpretation, and how to apply symbolic math libraries like SymPy for graphing & computation.
freecodecamp.org/news/learn-…
"Apologies if my emotions
And my feelings are all over the place
In my memory, that’s where you are, yeah
You’re my medicine, my ritual, yeah
And I will never be the same without you, my baby o
And I feel sick when I'm away from you, baby o"
