omniscient

[omniscient @omnisc1entt]

This is f*cking insane. Anthropic engineer:

"99% of people use Claude Code like Google. I run 100+ agents in a loop, with a chief agent and PM agents that manage the whole team"

In this 27-minute workshop, he shows how to get the most out of Claude Code for the least money

One Chat → Chief Agent → PM Agents → 100+ Workers → A Loop That Runs Itself

The 1% aren't smarter. They just stopped typing and started building the team

Worth more than another $500 vibe-coding course

Bookmark and watch it today

Then read how an AI employee runs QC for motion engineering with Opus 5.5 ↓

[m4lka @m4lkxx]

I fixed the same bug in two films

Opus never remembers my fixes

So I hired an AI employee to remember them for me, then tried to make him break my one rule on deadline

He didn't ↓ https://t.co/i7RRo3bIg2

[omniscient @omnisc1entt]

This is f*cking insane. Palantir CEO Alex Karp on what AI labs really want when they ask for regulation:

"If you build a technology that can destroy 10% of the world, that has civil and criminal liability attached to it

The only way to deal with that liability is to go to the government and say: nationalize us, please"

3.5M people watched this. Most still don't know how the technology works. Fix that, free:

Karpathy's Neural Networks: Zero to Hero
https://t.co/zwbDVAcleb

Hugging Face LLM course
https://t.co/sZShqipwyZ

Practical deep learning
https://t.co/ZbY7Aj50MN

Google's machine learning crash course
https://t.co/GJTOLoZH6V

Building effective agents
https://t.co/aAOxKV9oOl

Grok API docs
https://t.co/syPO7OEvaw

All of it costs $0

Bookmark it

Then read the "Harness, Loops, Graphs" guide below ↓

[omniscient @omnisc1entt]

This is f*cking insane. Palantir CEO Alex Karp on what AI labs really want when they ask for regulation:

"If you build a technology that can destroy 10% of the world, that has civil and criminal liability attached to it

The only way to deal with that liability is to go to the government and say: nationalize us, please"

3.5M people watched this. Most still don't know how the technology works. Fix that, free:

Karpathy's Neural Networks: Zero to Hero
https://t.co/zwbDVAcleb

Hugging Face LLM course
https://t.co/sZShqipwyZ

Practical deep learning
https://t.co/ZbY7Aj50MN

Google's machine learning crash course
https://t.co/GJTOLoZH6V

Building effective agents
https://t.co/aAOxKV9oOl

Grok API docs
https://t.co/syPO7OEvaw

All of it costs $0

Bookmark it

Then read the "Harness, Loops, Graphs" guide below ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. Geoffrey Hinton, the Godfather of AI:

"Your brain has a hundred trillion connections. A big language model has about 1% of that

And it still knows more than you"

In 90 seconds, he explains why, and it's not comforting

100 Trillion Connections → 2 Billion Seconds of Life → 1% of the Wiring → More Knowledge

The machine learns faster than we live

Bookmark it

Then read the "Harness, Loops, Graphs" guide below ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. The head of Claude Code at Anthropic:

"I'm not prompting my agents anymore

I'm building graphs and loops so they can build the agents for me"

In 11 minutes, he shows exactly where agent engineering is going next

Prompts → Agents → Loops → Graphs → Agents That Build Agents

The people who build the graph won't need to write the prompts

Worth more than most $1,000 agent courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below ↓

[omniscient @omnisc1entt]

This is f*cking insane. The head of Claude Code at Anthropic:

"I'm not prompting my agents anymore

I'm building graphs and loops so they can build the agents for me"

In 11 minutes, he shows exactly where agent engineering is going next

Prompts → Agents → Loops → Graphs → Agents That Build Agents

The people who build the graph won't need to write the prompts

Worth more than most $1,000 agent courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

The CEO of Anthropic and the CEO of Meta, in the same room, on camera, on whether AI is safe

Dario Amodei builds the model he says needs guardrails. Mark Zuckerberg gives his models away for free

4.6 minutes, two opposite bets on the most powerful technology ever built

Whose side are you on?

Then read the "Harness, Loops, Graphs" guide below ↓

[omniscient @omnisc1entt]

The CEO of Anthropic and the CEO of Meta, in the same room, on camera, on whether AI is safe

Dario Amodei builds the model he says needs guardrails. Mark Zuckerberg gives his models away for free

4.6 minutes, two opposite bets on the most powerful technology ever built

Whose side are you on?

Then read the "Harness, Loops, Graphs" guide below ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. Anthropic CEO Dario Amodei:

"There are jobs that took generations to build that may disappear"

High GDP growth and high unemployment at the same time. Software becoming essentially free to build

He's not predicting it. He says it's already happening

In this 31-minute interview, he explains how to prepare for what's coming

The gap isn't smart vs dumb anymore. It's people who use AI vs people who don't

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below ↓

[omniscient @omnisc1entt]

This is f*cking insane. Anthropic CEO Dario Amodei:

"There are jobs that took generations to build that may disappear"

High GDP growth and high unemployment at the same time. Software becoming essentially free to build

He's not predicting it. He says it's already happening

In this 31-minute interview, he explains how to prepare for what's coming

The gap isn't smart vs dumb anymore. It's people who use AI vs people who don't

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. Andrej Karpathy:

"Prompting is going away. Delete everything, keep the graph"

In 69 minutes, he shows how to build graphs of agents and why they're the only thing left standing

Prompts → Agents → Loops → Graphs

A loop keeps one agent working until the job is done. A graph decides which agents exist and what each one hands to the next

Worth more than most $1,000 agent courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below - graphs are the last layer ↓

[omniscient @omnisc1entt]

This is f*cking insane. Andrej Karpathy:

"Prompting is going away. Delete everything, keep the graph"

In 69 minutes, he shows how to build graphs of agents and why they're the only thing left standing

Prompts → Agents → Loops → Graphs

A loop keeps one agent working until the job is done. A graph decides which agents exist and what each one hands to the next

Worth more than most $1,000 agent courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below - graphs are the last layer ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. Anthropic engineer:

"You're not supposed to prompt Opus 5.5 anymore

90% of our engineers already built systems that prompt themselves. They fix their own mistakes and get sharper with every run"

In just 30 minutes, she builds one from scratch, live

Prompt → Memory → Loops → Self-Correction → Self-Improving Agent

Everyone has the same model. The system around it is the edge

Worth more than most $1,500 agent engineering courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below - memory is where it starts ↓

[omniscient @omnisc1entt]

This is f*cking insane. Anthropic engineer:

"You're not supposed to prompt Opus 5.5 anymore

90% of our engineers already built systems that prompt themselves. They fix their own mistakes and get sharper with every run"

In just 30 minutes, she builds one from scratch, live

Prompt → Memory → Loops → Self-Correction → Self-Improving Agent

Everyone has the same model. The system around it is the edge

Worth more than most $1,500 agent engineering courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below - memory is where it starts ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. Andrej Karpathy:

"The biggest mistake in AI right now is forcing agents to work before you master the harness around the model

We made that mistake at OpenAI in 2016. It cost us 5 years"

In this talk, he explains why the agent is not the product. The foundation is

Model → Harness → Loops → Agents That Emerge on Their Own

Demos are easy. Products take a decade

Worth more than most $1,000 agent courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below and build the foundation first ↓

[omniscient @omnisc1entt]

This is f*cking insane. Andrej Karpathy:

"The biggest mistake in AI right now is forcing agents to work before you master the harness around the model

We made that mistake at OpenAI in 2016. It cost us 5 years"

In this talk, he explains why the agent is not the product. The foundation is

Model → Harness → Loops → Agents That Emerge on Their Own

Demos are easy. Products take a decade

Worth more than most $1,000 agent courses

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below and build the foundation first ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. Andrej Karpathy:

"Vibe coding was fun. Harness engineering with Opus 5.5 is the next level

99% of my coding routine now runs on agents"

In a 30-minute talk, he shows how he built that workflow from scratch

Vibe Coding → Agents → Harness → Loops → Code That Ships Itself

Most people are still pasting code into a chat window

Worth more than any $1,000 agentic engineering course

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below and build the harness he's talking about ↓

[omniscient @omnisc1entt]

@a1254766624 just read the article below. why didn’t you do that before commenting?

[omniscient @omnisc1entt]

This is f*cking insane. Andrej Karpathy:

"Vibe coding was fun. Harness engineering with Opus 5.5 is the next level

99% of my coding routine now runs on agents"

In a 30-minute talk, he shows how he built that workflow from scratch

Vibe Coding → Agents → Harness → Loops → Code That Ships Itself

Most people are still pasting code into a chat window

Worth more than any $1,000 agentic engineering course

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below and build the harness he's talking about ↓

[m4lka @m4lkxx]

Your agent is not autonomous

You are its loop

You check every step because nothing else in the system can

I broke down the 7 levels from babysitting agents to running a fleet ↓ https://t.co/4RacdXA4nE

[omniscient @omnisc1entt]

This is f*cking insane. Anthropic engineer:

"Stop prompting Opus 5.5

Build a system that prompts itself. 90% of engineers at Anthropic already have one"

In 30 minutes she builds agents that remember, fix their own errors and get sharper every run

Prompt → Memory → Self-Check → Retry → Smarter Next Run

Worth more than any $2,000 agent course

Bookmark and watch it today

Then read the "Harness, Loops, Graphs" guide below and build the system, not the prompt ↓