Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Anthropic engineer: "You don't need better prompts. You need graph engineering: memory that stays, so your agent never forgets anything." In 28 minutes he shows what Anthropic does differently, how to build and structure work with agents. This beats any paid agent course I've seen. Watch it, then read...

466,014 Aufrufe • vor 3 Monaten •via X (Twitter)

37 Kommentare

Profilbild von darkzodchi
darkzodchivor 3 Monaten

original link:

Profilbild von Bacon armour
Bacon armourvor 3 Monaten

Like, I like liking, like likes like likers. But like, this guy like, says like, like a lot.

Profilbild von Stayu Kasabov
Stayu Kasabovvor 3 Monaten

It is getting ridiculous there. What's next buzzword? You need cosmic rays? There is nothing fundamentally different than using LLMs with scripts and data. :)

Profilbild von Michael Parenti
Michael Parentivor 3 Monaten

The problem is real. I live it every day. The paper’s diagnosis is exactly right. The prescription, for a software project, is where we part ways.

Profilbild von macintosh
macintoshvor 3 Monaten

I think I really need this

Profilbild von darkzodchi
darkzodchivor 3 Monaten

🙏🏼🙏🏼

Profilbild von Korens
Korensvor 3 Monaten

This is probably one of the most important shifts in agent design

Profilbild von SF
SFvor 3 Monaten

What does the talk have to do with graphs? I don t see it

Profilbild von Avid
Avidvor 3 Monaten

I think this is a good share

Profilbild von darkzodchi
darkzodchivor 3 Monaten

appreciate you brotha

Profilbild von BrianMcGrath
BrianMcGrathvor 3 Monaten

We need tools to better clean memories. My Claude memory is a hot mess. Would be nice to have tools to prune memory more selectively,

Profilbild von Vlad Oreshkov
Vlad Oreshkovvor 3 Monaten

We can now build this ourselves right? Into our own AI agents? I already have the obsidian second brain installed and created, maybe this would boost my agent’s memory even further.

Profilbild von Blum
Blumvor 3 Monaten

graph engineering is definitely worth diving deep into rn, imo

Profilbild von darkzodchi
darkzodchivor 3 Monaten

totally best time investement

Profilbild von Johnny Suede
Johnny Suedevor 3 Monaten

Memory that stays is half of it. The other half is that a graph gives you somewhere to put reconciliation. Sixteen agents can each remember perfectly and still hand you sixteen answers. Something has to decide which survives.

Profilbild von Web4 News
Web4 Newsvor 3 Monaten

The next AI breakthrough may come from better systems around models, not just better models. I read about it in @TheUpsideAI this morning

Profilbild von Gaetan Semet
Gaetan Semetvor 3 Monaten

We do not need bigger models. We need cheaper models. Thanks for your graph engineering, loopsy loopsa, I am pretty happy with my workflow. It works amazingly and is 10x more efficient than without it. Now I need cheaper Sonnet-level models. CHEAPER like less expensive.

Profilbild von Kanyasi Bence /Drevin.io
Kanyasi Bence /Drevin.iovor 3 Monaten

never forgets anything is the wrong target. a memory that keeps everything is a memory with no ranking, and retrieval quality drops as it grows. what you want is a system that forgets on purpose and can tell you why it kept the rest.

Profilbild von Matt
Mattvor 3 Monaten

Coding harnesses shouldn’t have memory

Profilbild von ヤバイ
ヤバイvor 3 Monaten

this video was dog shit

Profilbild von Dev老K-合约实盘
Dev老K-合约实盘vor 3 Monaten

@grok 总结下这个视频

Profilbild von Jodi Frank
Jodi Frankvor 3 Monaten

This would be an amazing improvement!

Profilbild von Pearl AI
Pearl AIvor 3 Monaten

the next AI breakthrough might be better memory, not better prompts

Profilbild von darkzodchi
darkzodchivor 3 Monaten

exactly

Profilbild von Shivam
Shivamvor 3 Monaten

You still need better prompts.

Profilbild von Harley Lewis Foote
Harley Lewis Footevor 3 Monaten

Memory that never forgets anything.

Profilbild von Gerard Sans | Axiom 🇬🇧
Gerard Sans | Axiom 🇬🇧vor 3 Monaten

There will be signs. The signs:

Profilbild von Johnny Utah
Johnny Utahvor 3 Monaten

It’s mostly a guide on: ✅ Prompt caching ✅ Separating ingestion from retrieval ✅ Low effort for extraction, high effort for reasoning ✅ Batch processing historical data Useful engineering advice, just don’t expect a brand new graph memory architecture inside.

Profilbild von Magicfit
Magicfitvor 3 Monaten

memory beats prompting once agents get real

Profilbild von Akbar Shaik
Akbar Shaikvor 3 Monaten

One thing I really like about this direction is that it shifts the conversation away from "better prompts" and toward better system design.

Profilbild von Pham Gia Trang
Pham Gia Trangvor 3 Monaten

The evolution of cybersecurity is shaped by breakthroughs in research, market demand, and infrastructure maturity in emerging economies.

Profilbild von Philipp
Philippvor 3 Monaten

Actually the graph engineering is basically the Langraph implementation of nodes

Profilbild von Steve Markbury
Steve Markburyvor 3 Monaten

Nice

Profilbild von Uttam Gupta
Uttam Guptavor 3 Monaten

Better prompts help once. Better memory compounds across every interaction. That’s the real shift from using AI as a chatbot to building agents that improve with context.

Profilbild von Harshil Tomar
Harshil Tomarvor 3 Monaten

pretty insightful

Profilbild von Yel
Yelvor 3 Monaten

Please sell me your 10x graph engineering course made with leaked $2.2M Anthropic engineer scripts.

Profilbild von Formulate AI
Formulate AIvor 3 Monaten

Future

Ähnliche Videos