Anthropic engineer at Stanford: “80% of our engineers are using self-improving loops. Now everyone is building agentic graphs.” in 1-hour Stanford lecture, Anthropic engineer reveals how Claude thinks - and why Graphs are the future. This 1-hour watch will replace your $500 course on agentic engineering. Watch it today,... then read how to become a Graph Engineer in the article below.show more

what does claude graph architecture look like when multiple self-improving loops share state safely?

thanks, dude

worth blocking the full hour for this one

Yeah bro. This is limited Anthropic lecture at Stanford

Graph based architectures keep unlocking smarter reasoning and better automation every month

Yeah. But also Graphs are important for agentic memory

i'm seeing this person for the first time, but this is definitely something worth watching

Yeah. Amazing watch bro. Anthropic at Stanford

appreciate the share Codez loved the graphs angle but ngl self improving loops scare me, every one i ran slowly turned into something i couldnt debug :(

After this 1-hour lecture, I’m sure you will change your mind about building loops and graphs!

agents need explicit coordination

Yeah. Graphs is the best layer for everything

Intriguing glimpse into AI development and future trends.

80% adoption for self-improving loops sounds high but tracks with where tooling is heading. The graph architecture part is the interesting bit though.

It takes longer to implement this sort of stuff than for a better method to develop

Just finished the full Stanford lecture — the computational graphs section is pure gold. Graphs really do feel like the next leap for agentic systems.

will check full video tonight mate, thx for sharing

That is wild! 🤯 When 80% of the engineers at a top AI lab are using self-improving loops, you know agentic graphs are the real deal, not just a passing trend.

자꾸 새로운게 나오네. 조만간 해봐야겠다

Why people isn't learning this man

this is a lot to take in, but can it really change the game?

graphs finally give agents real memory + coordination instead of just fancy chains. this lecture makes it click hard.

Realmente esta charla me hizo replantearme muchas cosas

let me watching this lecture

moving past rigid multi-step pipelines to agentic graphs is the only way to scale code workflows