正在加载视频...

视频加载失败

Google just released 1-hour workshop on "Graph Engineering" for agentic systems from 0% to 100%: "loop agent is getting deprecated soon, so I’m going to focus on graph workflows" 0% → 8:01 - build your first agent from scratch 30% → 24:14 - Prompt engineering 55% → 44:48 -...

22,815 次观看 • 1 个月前 •via X (Twitter)

15 条评论

Argona 的头像
Argona1 个月前

most of my wins came from deleting agents

codila 的头像
codila1 个月前

hahah, deleting bad agents you wanted say

Argona 的头像
Argona1 个月前

yeah the redundant summarizer and reviewer ones went first. how much latency did you claw back swapping them for plain functions?

chuplung 的头像
chuplung1 个月前

solid breakdown saving this

codila 的头像
codila1 个月前

greatest decision for today!

0xRafy 的头像
0xRafy1 个月前

love these google courses, always useful for my job

codila 的头像
codila1 个月前

true! last week i watch maybe 2 or 3 these courses and bro, it's real practice info

vartekx 的头像
vartekx1 个月前

great info bro, saved for personal use

codila 的头像
codila1 个月前

just copypast this post into your chat and job is done

newry 的头像
newry1 个月前

Does the video explain what Graph Engineering is?

codila 的头像
codila1 个月前

exactly and how to build it from scratch

Kravn 的头像
Kravn1 个月前

router without a model call idea is genuinely practical - saves real tokens on simple brand decisions

Siam 的头像
Siam1 个月前

yeah I agree. graph workflows just seem to make more sense once agents start doing stuff in parallel.

Fajar M Reza 的头像
Fajar M Reza1 个月前

Which failure cases does graph engineering handle better than linear agent workflows?

AI Mastery Guide 的头像
AI Mastery Guide1 个月前

Deletes three agents, bold 👀

相关视频