Loading video...

Video Failed to Load

Go Home

Ex-Google Jeff Dean just released the best 1-hour lecture on AI engineering: from basics to Graphs 1:45 - LLM from scratch 17:22 - How to use AI models 30:03 - Prompt engineering 52:35 - One human coordinating 100 agents 27 years of building AI at Google, compressed into one...

344,075 views • 6 days ago •via X (Twitter)

16 Comments

Je Je's profile picture
Je Je5 days ago

Here is the Youtube video if someone is interested:

Roshni's profile picture
Roshni6 days ago

27 years of experience condensed into an hour... this is an absolute must-watch for any AI engineer. 🧠

Kshitij Mishra | AI & Tech's profile picture
Kshitij Mishra | AI & Tech6 days ago

keep posting more

rewind's profile picture
rewind6 days ago

hundred agents one human, sure

Raunak Yadush's profile picture
Raunak Yadush5 days ago

One hour covering everything from LLMs to 100 AI agents.That’s a serious masterclass.

David Starmac Ai's profile picture
David Starmac Ai6 days ago

The 52:35 part on one human coordinating 100 agents is the one people will skip. In practice the hard part isn't coordination, it's deciding when to trust the output.

Kuzka_aaa's profile picture
Kuzka_aaa6 days ago

The useful cut is the last chapter.LLM from scratch and prompting are table stakes. The hour only gets interesting at 52:35 — one human coordinating 100 agents.That’s the job now: orchestration, not another prompt trick.

Alex's profile picture
Alex6 days ago

jeff dean's a genius, love watching his talks

Morty's profile picture
Morty6 days ago

one of my fav video with him, gonna watch it again

AI Mastery Guide's profile picture
AI Mastery Guide6 days ago

27 years in one video, saving this

Anra's profile picture
Anra6 days ago

context engineering beat prompt engineering as a term because the input isn't just words anymore — retrieval order, tool results, session state, timing. the prompt is the smallest part.

null's profile picture
null6 days ago

any slides mate ?

brnrd's profile picture
brnrd6 days ago

The jump from “how to prompt a model” to “one human coordinating 100 agents” is the whole story. At 100 agents, the hard problem isn’t prompting anymore 🤓 It’s state, ownership, delegation, verification, authority, recovery - and giving one human a control surface that doesn’t turn them into the message bus. That’s where agent engineering becomes runtime engineering.

Collins's profile picture
Collins6 days ago

AI engineering is shifting from building models to orchestrating intelligence

qiuqiang's profile picture
qiuqiang4 days ago

nice

Vivek Yaligar's profile picture
Vivek Yaligar6 days ago

Once one person can coordinate 100 agents, the bottleneck isn’t throughput — it’s decision rights and quality bars. Who sets the definition of done, who catches silent failure, who owns the call when agents disagree. That’s the new operating system.

Related Videos