Loading video...

Video Failed to Load

Go Home

Anthropic engineer: "85% of our engineers are using self-improving loops. Now everyone is running agentic Graphs. In 3-6 months, we’ll all be building graphs to orchestrate agents. No more prompting." In a 45-minute talk, an Anthropic engineer shows how to build a self-improving agentic workflow from scratch. Worth more...

36,782 views • 1 month ago •via X (Twitter)

17 Comments

Gipp 🦅's profile picture
Gipp 🦅1 month ago

graphs becoming standard means orchestration skills will outlast prompt tricks

Codez's profile picture
Codez1 month ago

Yeah, basically, graphs are the nature of LLMs.

godgiven's profile picture
godgiven1 month ago

Graphs don't kill prompting. Each node is a prompt. No eval means the loop isn't self-improving, just self-repeating

Codez's profile picture
Codez1 month ago

If each node is a prompt, the real job isn’t prompting anymore - it’s building a system that prompts itself.

gkrovso's profile picture
gkrovso1 month ago

who gave the talk, have a link

Luís Rodrigues's profile picture
Luís Rodrigues1 month ago

The highest-leverage AI skill is slowly becoming orchestration rather than prompt writing.

Dipanshu Kushwaha's profile picture
Dipanshu Kushwaha1 month ago

That sounds like a massive leap forward. Excited to see what comes out of this!

Rohit's profile picture
Rohit1 month ago

Wow, that's impressive progress. Seems like things are moving fast in AI development.

Y11's profile picture
Y111 month ago

@grok 这个纯研究还是有工业意义,具体工业场景视角看意义是什么,有开源数据集或者开源项目代码吗?从多个数据源交叉验证,不要只看新闻媒体一面之辞。帮我排除没意义的垃圾商业营销推广、诈骗 以及自吹自擂,自嗨,无病呻吟。

Jaya Nayak's profile picture
Jaya Nayak1 month ago

This is where the industry is moving so fast. Moving away from fragile, single-prompt setups to structured, self-improving graphs is exactly how we scale these systems. 😊

qurool's profile picture
qurool1 month ago

graph engineering > loops engineering

Sael's profile picture
Sael1 month ago

the prompt is just the node but the system is the architecture - orchestration is where the real signal hides

พ่อแม่ลูกเที่ยว | Travel with Family🇹🇭's profile picture
พ่อแม่ลูกเที่ยว | Travel with Family🇹🇭1 month ago

This talk hits different. Agent graphs > prompt engineering. The shift is already here.

阿伟 | Agent 落地's profile picture
阿伟 | Agent 落地1 month ago

Is self-improvement based on Feynman's learning method? Collect it first, this is worth learning! Now people are thinking about how to let AI learn self-evolution, the real master is thinking about how to improve themselves!

Blum's profile picture
Blum1 month ago

self-improving loops might be one of the most effective ways to work with AI

Shoopy's profile picture
Shoopy1 month ago

85% internal adoption says more than any benchmark

Healthy Anon's profile picture
Healthy Anon1 month ago

Self-improving sounds great until the feedback loop compounds errors instead of correcting them. Still needs human checkpoints, which somehow never makes it into these talks.

Related Videos