Andrej Karpathy could have charged $10,000 for this course. He put it on YouTube. The man who built Tesla Autopilot from scratch. Co-founded OpenAI. Understands AI at a level most engineers at Google and Meta never reach. Sat down. Recorded 2 hours. No frameworks. No libraries. No shortcuts. Then... dropped it for free. The gap between people who watch it this week and those who save it for later is not 2 hours. It is everything those 2 hours quietly unlock for the rest of your career.show more

- from 3 years ago

Karpathy has done roughly one of these every year since 2015. The cs231n lectures, the RNN essays, Micrograd, the nanoGPT walkthroughs. Each one blows up the week it drops and then sits quietly in thousands of 'watch later' queues. His time was never the bottleneck here.

i am 1 hour in broski i dont think ama gonna make it

@chuchunav_eth i believe you can bro!

He has enough money to do everything for free, forever. Many of us aren’t there yet…

Source

nice so you downloaded it and re-uploaded it to farm engagement

@grok how old is this video and what is the direct link for it on YouTube ?

Here's the full Zero to Hero course from @karpathy, but interactive

The frameworks part hits hard. Spent years debugging PyTorch configs when raw NumPy would've taught me 10x more about what's actually happening under the hood. Karpathy gets that the fundamentals matter more than the tooling.

There are so many posts with a similar vibe of “watch this video, new AI update here, new Claude version is out — it’s game changing. If you don’t keep up, you’ll be left behind.” It’s always the same pitch with the same FOMO grab. These posts are all the same.

He’s truly a legend for everything that he’s done in the space

huge legend!

the thing people miss is that andrej isn't just sharing knowledge, he's showing how to think about problems you didn't even know existed yet. that's the value in his approach, not the code itself.

This playlist is pure gold for anyone serious about actually understanding the fundamentals.

@grok do I need to know this as a non-technical who dabbles with Claude cowork and has been regular AI user

@cyrilXBT

What’s the YouTube link please?

The "save for later" folder is a graveyard. Watched it today instead.

It's a nice way to show how the rationale behind neural nets are. While taking TTIC 31230 back in 2018, David McAllester used a same approach with his EDF (Educational Framework):

UNLOCK THE REST OF YOUR CAREER FOR FREE Invest 2 hours of your time. Don't save for later. Andrej Karpathy could have charged $10k for this

Awesome post @neil_xbt

I think he is one of the only genius people in today's world that wants to teach his findings to others

could have charged $10k is speculative lol, educational content market doesn't support that pricing unless it includes certification job placement or direct mentorship, free youtube vs paid course aren't substitutes

Absolutely worth watching for anyone interested in how AI works/is trained.

genuinely one of those people where you finish watching something and feel like your brain got quietly reorganized. hard to explain until you've experienced it

@grok source link?

Same. His clarity is unreal — like watching someone untangle spaghetti code in real time. I’ve rewatched that first lecture 3x while building DangcingAI’s dubbing pipeline… still catching new insights. What part hit you hardest?

How useful is this @grok?

Karpathy is a boss do doubt

👇

could have pasted the utube link dawg ngl

Isn’t this video like a year old? lol

Most people won’t realize what they’re sitting on

Drop your micrograd version / fork please.

@grok what is the video about?

summarize this into one paragraph and tell me the main learning points. dont mess up "