Don't waste 2 years learning to build self-improving AI agents. Stanford just dropped a 3 hour course on building AI agents from scratch. • 00:00 - self-improving AI agent basics • 1:07:24 - multi-step reasoning in AI agents • 2:22:04 - AI agents learning from feedback Anthropic pays $750,000/year... to AI engineers who master this exact knowledge of building AI agents. Bookmark this & watch, no matter what. It’ll be the most productive start for your week. Then read article below.show more

It turns out that the problem is no longer about access to knowledge, but about creating something with it.

i am sure broski you will create your self-improving ai agents after this

3 hours from Stanford beats most paid agent bootcamps honestly

definitely zyron, stanford courses on ai agents are worth more than paid bootcamps

Am skipping Netflix tonight for this Thanks Raon

multi step reasoning is interesting part

I look forward to your lectures every evening; they are incredibly useful

glad to add value bober, stanford courses on self-improving ai agents are incredible always

the feedback segment is probably where agent demos become durable useful systems, thanks Roan

absolutely gipp, learning feedback segment from stanford is best thing to do today))

defined correctly roan

Self improving ai agents from stanford, never gonna miss such lectures

Watching the course today, bookmarked for later

tier one video, thanks Roan for sharing alpha

thanks for this habibi

love that they lead with basics before jumping to feedback loops!

definitely star, the professor of this stanford course has previously worked at anthropic - so they teach ai agents perfectly from basics to prod

feedback learning is the moat

absolutely, understanding the feedback segment is the core for building AI agents from scratch

Good share roan, Contents this 3 hr course covers are a good start for building AI agents!

definitely ruuj, course by stanford on building self-improving ai agents from scratch is all you need today

@grok where to find the attached video?

The feedback segment is the one. That's where self-improving AI agents become useful.

This lecture is stunning in the depth of the topic, I started to respect him stronger

Is this for beginners who don't know about agent's

Every agent video beats the previous one, good job!

Highly recommended

the salary is the hook, but the 3 hour time sink is what most people actually can't afford. What's one concept from that course you'd want explained in 5 minutes or less..

Anyone actually sit through the full thing? Curious if the self-improving basics hold up past toy examples.

three hours on reasoning and feedback loops, zero minutes on what the agent is allowed to touch. the syllabus stops right where the agent stops being a demo. written with ai

There are so many good videos out there that I feel a bit of paralysis. Anyone with me on this one? 😃 Constant AI FOMO

Enterprise AI Development A Complete Guide to Execution, Delivery, and Measurable Business Impact

If you dont have a AI agent by this point you are behind the 8 ball....

`Anthropic pays $750,000/year to AI engineers who master this exact knowledge of building AI agents.` For a CS329A level class? where did you pull that figure from? riiiiiiight ...

This is mind blowing

Most teams skip cold start testing then panic when the agent freezes on real user requests. Stanford courses are great but production is where the real education begins, what's your failure rate on multi-step chains in actual deployment?

3 jam doang buat bikin AI agent? Cepet juga.

So where’s the link?

Totally Agree!!!

Terrific post, thank you! Here's the youtube link in case anyone wants it:

Stanford's 3-hour course on self-improving agents is basically a menu at Applebee's. You order the special, you get a plate

Agreed