Anthropic pays $750,000+ a year for engineers who can build LLM architectures from scratch.
Stanford taught the entire thing in 1 hour lecture & released it for free.
Bookmark this before it gets buried.
CapCut, but in your browser — no installs, no watermarks.
Shipped this week to OpenReel:
- Drag-and-drop effects & transitions
- 3D text & objects + track mattes
- Speed-curve ramps
- Redesigned tabbed editor + compact mode
- Smooth playback that follows your playhead
https://t.co/kiEseCJL1e
March 2023. Meta's LLaMA leaks.
Suddenly everyone wants to run it ~ and hits one wall. You need a $5,000 GPU to even turn it on. 🤯
One guy in Bulgaria looks at it and goes…nah.
Meet Georgi Gerganov 🇧🇬
> No team. No funding. No San Francisco.
> Just a developer quietly building a small C library in his spare time.
> While everyone else waited for GPUs, he tried to run LLaMA on his MacBook.
> One evening. Pure C/C++. No Python. No CUDA. No cloud.
> His own README said: "hacked in an evening, I have no idea if it works."
> It worked.
> He called it llama.cpp. 🚀
> Within days people were running real AI on laptops, Raspberry Pis, phones.
> The thing the whole industry said was impossible ~ one guy did it in a single evening.
> He'd already done the same earlier with whisper.cpp for voice.
> Today every local AI tool you've heard of ~ Ollama, LM Studio, GPT4All ~ runs on his code.
> 114k+ GitHub stars.
> Feb 20, 2026, Hugging Face acquired ggml. ai (his company/team). Code stays 100% open-source and free.
Most engineers spend years trying to get noticed.
This guy spent one evening, called his own work broken, and started the local-AI era.
What a man. 🐐
Never in my wildest dreams have I thought that I would see books, written in Japanese, in a book store at the other side of the world, containing a whole chapter about the tools I build and maintain every single day. This is truly amazing ❤️
New article: a visual tour of recent LLM architecture advances, from Gemma 4 to DeepSeek V4.
I focus on long-context efficiency tweaks like KV sharing, per-layer embeddings, layer-wise attention budgets, compressed attention, and mHC.
Link: https://t.co/KO81y3kTH7
@NITAGhana
I'm not sure why you're so intent on coming for the tech industry.
Now you want to restrict private organizations from hiring us unless we get a government certification?
What have we done to you? Seriously.
You keep doing this. It simply cannot be accidental.
A moment of stillness by the water, with Chongqing’s glittering skyline as the backdrop, blending the city’s vibrant energy with a gentle evening calm.🤍🗺️✨
#Chongqing#Cityscape#Riverside#UrbanView
📸langgejiaoniyongshoujipairenxiang
I still keep saying this. We handed over a site to Huawei and they put their product there while we worked with them in parallel for 2 years. In those two years, there was ZERO downtime on their platform because the Chinese were ALWAYS working and proactive. They never rest. To compete, you must have the same work ethic.
SNAPCHAT PAID $150,000,000 FOR LOOKSERY - STARTUP IN DEEP LEARNING COMPUTER VISION
This 1-hour Stanford lecture on "DL for Computer Vision" will teach you how to build same project from scratch.
Bookmark it & watch today. Stanford's full course (19 lectures) on Deep Learning for Computer Vision below ↓
Atlassian just reported $1.79B in quarterly revenue and serves 350,000+ customers
then fired the engineer who built their infrastructure
He shares the whole thing
a breakdown of Atlassian’s playbook:
> Envoy over enterprise load balancers > sidecars for auth + logging + rate limits > DynamoDB + SQS > automated VM deployments
GPU shortage is worse than ever.
H100s cost more today than they did 3 years ago, and you cannot get them on-demand.
The big AI labs have locked up most of the supply for years. I’m worried university researchers and individual developers simply won’t be able to get GPUs.