Loading video...

Video Failed to Load

Go Home

Google Engineers just dropped a 1-hour course on agentic engineering from scratch: 0% - 00:00 – How to build your first AI agent 25% - 08:24 – Build agent memory (short, persistent, long) 50% - 28:34 – Agentic loops, long-running AI agents 100% - 1:00:22 – Multi-agentic systems This...

12,070 views • 2 months ago •via X (Twitter)

16 Comments

Apex's profile picture
Apex2 months ago

Thanks chad, gonna watch it now

Ricker's profile picture
Ricker2 months ago

enjoy bro

Argona's profile picture
Argona2 months ago

saved it mine held up until two agents wrote the same file bro

Ricker's profile picture
Ricker2 months ago

glad to be useful

Pedro's profile picture
Pedro2 months ago

this is 1h gem

Ricker's profile picture
Ricker2 months ago

fact

Morty's profile picture
Morty2 months ago

This video delivers huge value honestly

Ricker's profile picture
Ricker2 months ago

exactly man

rewind's profile picture
rewind2 months ago

persistent memory is where it starts

Vǫlvabard's profile picture
Vǫlvabard2 months ago

@grok

Brice's profile picture
Brice2 months ago

Saving this to watch later

Carlos Matias Baglieri's profile picture
Carlos Matias Baglieri2 months ago

Solid syllabus — from formalizing agentic nodes myself, the part that bites is the envelope: a ReAct-style loop needs a well-founded progress measure to guarantee termination, and tool calls need confinement to an authorized capability set. Multi-agent systems amplif ❤️ #zypho_ai

Mark's profile picture
Mark2 months ago

Useful breakdown. Memory and loops are the practical parts operators need first.

morph's profile picture
morph2 months ago

Amazing vid for my dinner

Allan MacGregor's profile picture
Allan MacGregor2 months ago

ok, sounds like a good way to spend a Saturday morning while I drink my coffee

Fajar M Reza's profile picture
Fajar M Reza2 months ago

The LLM-to-graph progression shows orchestration becoming the real engineering bottleneck.

Related Videos