Raytar

Anthropic is paying $500k-$850k/year for engineers who can build AI agents. Two Anthropic engineers just gave away a 75-minute workshop: 27:01 - A full app from one prompt 37:55 - Anthropic's real agent setup 54:38 - Permanent memory for Claude 1:08:38 - Make Claude work while you sleep Prompting Claude is becoming basic. Make Claude prompt itself. Watch the workshop, then read the guide to building loops below.

4 Anthropic engineers. One hour. The best prompting lesson on the internet, and it's free. 14:12 - Real prompts from inside Anthropic 32:33 - The fix that saves your prompts 59:44 - How to give Claude the right context 1:07:59 - The prompt that does your job while you sleep Spend one hour with the people who write the prompts inside Anthropic. Then read the step-by-step guide below.

Anthropic engineer: «You're not supposed to sit and re-prompt Claude all day. You build one loop, and it does the work while you sleep.» • 00:00 - Build your first agent that runs on its own • 05:02 - The loop that fixes its own mistakes • 16:43 - Why Claude forgets you between chats, and the fix You can learn it in one afternoon. 30 minutes here replace what most people spend a year figuring out. Watch it today, then read the guide on building loops below.

@0xSlyth come back with your main takeaway, curious what hits

@Raytar watching this tomorrow first thing

Anthropic just released a 4-hour course to getting a $500k AI engineering job: 00:15 - The right way to prompt Claude 33:21 - What makes Claude act dumber on your code 01:33:39 - How Anthropic use Claude every day 02:50:56 - The fix that makes Claude way smarter This 4-hour Anthropic free course replaces about 10 paid engineering courses. Watch it today, then read the step-by-step guide on building loops below.

@Kevthewriter glad it landed! more in the works

@Raytar that's gold, ty for sharing mate

My coworker mass-applied to AI jobs for two years. No CS degree. No callbacks. Last month Anthropic offered him $630K. He watched one 28-minute talk. An Anthropic engineer. Called me after and said one thing: "I've been prompting Claude like an idiot this whole time." He doesn't write prompts anymore. He writes loops. The loops write the prompts. The talk and the full guide on loops are below.

Anthropic just dropped a 1-hour workshop on how to actually build with AI in 2026: 00:26 - Why they stopped prompting 16:57 - The system that builds, tests, and fixes itself 25:27 - One engineer replacing a 5-person team 36:25 - Building your first loop from scratch This 1-hour watch could replace 10 AI engineering courses on the internet. Watch it today, then read the step-by-step guide on building loops below.

Our junior failed mid. 3 months later he passed senior. One 96-minute video. He watched it 11 times. A team lead dropped it in our Slack back in March. A workshop recording. The guy shows how to make Claude work while you sleep. It also explains why Claude gets dumber halfway through every big task. The fix is at minute 9. $58k in March. $128k last week. The video was never published. I kept the mp4. It's below.

A senior IBM engineer explains Skills better than any $500 course. Free on YouTube. 12 minutes. No one has seen it. 0:36 - The gap skills fill 1:41 - What a skill actually is 4:34 - How agents handle 100 skills at once 6:28 - Skills vs MCP vs RAG vs fine-tuning 10:29 - A skill can hack your laptop 14 months I used Claude before this video. 14 wasted months. Watch it, then read the full guide about loops below.

Anthropic shared their video on a weird geek forum. Two Anthropic engineers explain how they actually prompt Claude. This is the best way you can spend 29 minutes. I watched it last night. Halfway through, I got angry at myself. Two years I've been using Claude wrong. First lecture in two years that actually changed how I work. Watch the video. Then read the article below.

An Anthropic engineer showed me why prompting isn't enough. 2:47am. He dropped a video in a Claude dev Discord. 36 minutes long. 290 views. He explains how to build a loop so your agent can fix itself and finish without you. Watched it at 3am. By 3:20 my terminal was open. By 3:40 I was already rewriting how I work. The video disappeared the next day. I kept the mp4. It's below.

Boris Cherny recorded this internally for his team. It was never meant to go public. He built Claude Code and shows how to go from idea to production-ready product in just 30 minutes. No script. No cuts. He just talks and types. This is the most practical video on Claude Code I’ve seen. Watch the video, then read about loops below.

It was a Zoom call. Somebody hit record. Thank god. Boris Cherny, who built Claude Code, said: «At Anthropic, loops review every pull request. Every single one. Give it 3-6 months and prompting is gone» Loops changed how I use Claude. I don't really prompt anymore. This recording is worth more than any $500 course on agents I've seen. Watch this video, read about loops below, then unfollow me. Fair trade

Google rejected my friend three times. One lecture later, they offered him $620k. This is the lecture where Karpathy shows how LLMs actually think, not how to prompt them. He called me right after and said: “This is the first time it actually made sense.” Three days later he applied again. Every question they asked about, he already knew from that video.

An ex-Google engineer just showed how to actually build AI agent loops. This is the clearest explanation I’ve seen. Here’s exactly how he does it: → Give it memory → Harness the loop → Trace what happened → Judge the result → Fix what’s broken → Ship I tried it the same night. The difference was immediate. Most people are still prompting agents like it’s 2024. Save this. Then watch the full explanation below.

Creator of Claude Code, Boris Cherny: «Prompting is dead. At Anthropic, we only use self-improving loops now» 20 minutes inside how the Claude team actually works with Claude every day. I watched 40+ videos about Claude this year. This is the only one I'd rewatch.

My friend launched a big as f*ck design agency last month that already runs completely on AI. In his first month he made $37k. All because of this video. Ad creatives that used to take 2 hours now take just 5-7 minutes. Now his entire team runs everything on Claude and loops. Claude creates the designs while loops manage everything with zero human input. Watch the video first. Then see how to build loops yourself.

@kepochnik thank you for the quote maaann

DeepMind agents engineer: "We define instructions, we run it, we observe, we adjust. We run it again." in 10 minutes, Philip breaks down why coding agents fail and what actually fixes it here's what he covers: > why agents need loops, not step-by-step instructions > treating errors as inputs instead of crashes > why unit tests don't work anymore, and what replaces them > designing APIs agents can actually understand on their own most engineers are still building agents like they built traditional software the ones getting reliable results stopped fighting the model and started looping full breakdown is in the article below👇

@Raytar thanks to you for the alpha article testing your style on your article

OpenAI just dropped something f*cking dangerous. my friend who works there and makes $1.2M a year sent me this video at 3am and said: " 00:33 will change your life forever." Sam Altman shows for the first time how his team actually uses ChatGPT every day for real work. I watched it and realized I've been using this completely wrong for two years. While I was asking questions - they were sleeping and AI was working for them. Watch the video below then read the article.