正在加载视频...
视频加载失败
Google just released a 1-hour course on “Context Engineering” for AI agents from scratch: • 00:00 - Introduction to “Context Engineering” • 03:02 - Context Engineering vs “prompting” • 16:18 - Build a context brain for your agent • 22:10 - Building agent context (tools, skills) • 29:06 -... show more
14,822 次观看 • 1 个月前 •via X (Twitter)
25 条评论

Just use ICM

every context course teaches you to add more unfortunately i've bought courses like that, and i'm sure this video covers most of the paid material

Google explains how to build context differently, bro.

good share

Thx Mayor !

The future belongs to agents with context, not just clever prompts

the future is agents building agents, that are building their context

the tools and skills section shows context is more than prompts

Yeah. This is true Gip. And it’s very important to work with Context correctly for long running AI systems

one hour could change workflows

Context engineering being a whole course now is wild, it was the invisible job a year ago. Even (building it) handles context structurally so agents arrive with the right state, not a wall of text.

So 'Context Engineering' is just what we’re calling 'Prompting' now that it has a LinkedIn profile?

Learning something is now very easy people just need to focus on t

Context Engineering change this AI-meta thank you

love this

Context engineering is definitely the next major high paying tech skill.

the memory bank section at is the one

One hour on context engineering means the gap wasn't tooling, it was people feeding the model wrong

Agents are great at refactors, but verification is the bottleneck. Instrumenting the whole pipeline — model, prompt, code diff — is the only way to audit results later. Tools like give you that trail.

free google drop on context engineering hits different, the memory bank part is what most paid courses skip

Solid resource. Context management is still the biggest bottleneck for most agent implementations I've seen in production.

This video has closed caption. Why can't Grok summarize this. There are hundreds of hour long talks. I don't have time for them. @elonmusk @grok @SpaceXAI @X

Context engineering is still under-discussed compared to prompt engineering. Most agent failures I see come from messy or incomplete context, not from the model itself.

It is great 1-hour course to watch, well done Movez

Excuse me Sir where is the original video? This is stitched together.
