the0xbt

learning AI by building agents | graph engineering & posting what actually works

[the0xbt @the0xbt]

Anthropic engineer:

"Most people are using 10% of what Claude can do.

The other 90% is Loops and Graphs".

In 25 minutes he explains how Anthropic actually uses Claude, and why most people never get there:

Chat -> Agents -> Loops -> Graphs

A loop closes one job without you and a graph decides which jobs exist at all.

Claude was never hard to use, loops and graphs are how you use it well..

Watch it today and then save the full graph engineering guide below before everyone catches up ↓

[the0xbt @the0xbt]

@slash1sol sure

[the0xbt @the0xbt]

Anthropic engineer:

"Most people are using 10% of what Claude can do.

The other 90% is Loops and Graphs".

In 25 minutes he explains how Anthropic actually uses Claude, and why most people never get there:

Chat -> Agents -> Loops -> Graphs

A loop closes one job without you and a graph decides which jobs exist at all.

Claude was never hard to use, loops and graphs are how you use it well..

Watch it today and then save the full graph engineering guide below before everyone catches up ↓

[slash1s @slash1sol]

https://t.co/fYpaaiH1dP

[the0xbt @the0xbt]

@virgilxbt me too

[the0xbt @the0xbt]

@virgilxbt worth to watch 100%

[the0xbt @the0xbt]

@itsthedonhashim fully agree bro yeah

[the0xbt @the0xbt]

Google just put out a free 1.5 hours workshop.

On taking AI systems from plain RAG to graph-based memory:

• 15:32 - Stand up the production agent stack.
• 28:00 - Turn scattered data into a knowledge graph.
• 41:00 - Build GraphRAG with semantic and hybrid search.
• 58:00 - Pull graph context out of images, text and video.
• 1:09:00 - Orchestrate specialized agents with Google ADK.
• 1:20:00 - Give agents memory that survives the session.

Most teams still stop at RAG and Google is already teaching what comes after it:

RAG -> GraphRAG -> Memory -> Multimodal Agent Graphs

Built on Spanner Graph, Gemini, ADK and Memory Bank.

This free workshop beats most paid GraphRAG courses.

Watch it today, then read the full From RAG to Context Graphs roadmap below ↓

[slash1s @slash1sol]

https://t.co/fYpaaiH1dP

[the0xbt @the0xbt]

@gippp69 yeah mate

[the0xbt @the0xbt]

@0xMfox sadly but yes..

[the0xbt @the0xbt]

@notp3rk yep that's the point bro

[the0xbt @the0xbt]

@virgilxbt 👀

[the0xbt @the0xbt]

@slash1sol let's go

[the0xbt @the0xbt]

@virgilxbt interesting to watch it today

[the0xbt @the0xbt]

@AleiahLock and 100% useful bro

[the0xbt @the0xbt]

@Lummox_eth agree bro

[the0xbt @the0xbt]

A Chinese developer explained why Kimi K3 beats bigger models better than anyone in English:

"Many people are waiting for the next disruptive technology.

The real gap is usually this: the right direction, and every detail done the right way".

In 22 minutes he walks through the three axes Moonshot scaled at once.

Tokens -> Context -> Agents -> Graphs

Muon doubles token efficiency against AdamW, and Moonshot were the first to prove it holds at a trillion parameters.

The third axis is the one nobody covers -- the agent swarm stacked on top of the other two.

This one breakdown covers more than most $1,000 agent engineering courses.

Watch it now, then lock in the full A-Z guide below ↓

[slash1s @slash1sol]

https://t.co/fYpaaiH1dP

[the0xbt @the0xbt]

@virgilxbt sounds good 👀

[the0xbt @the0xbt]

@BaseUPgame yeaaah

[the0xbt @the0xbt]

@Atenov_D yep this is true

[the0xbt @the0xbt]

Anthropic Research Lead:

"99% of our engineers run swarms of 300+ self-improving agents.

Now everyone is building agentic graphs".

Agents -> Loops -> Graphs -> Self-Improving Systems

In 20 minutes the session breaks down how graph engineering turns isolated agents into systems that improve themselves.

The real setup is Claude running through graph workflows, plan mode and dynamic orchestration.

Priced at $700 this would sell as a graph engineering course, and it costs nothing.

Bookmark it, watch the talk, then read the article below ↓

[slash1s @slash1sol]

https://t.co/fYpaaiH1dP