Hey Reader! I poked a few Model Context Protocol servers this week and stumbled on one with zero security. If you’re using (or about to use) MCP, watch this: 👉 Hacking MCP: We Found a Server with ZERO Protection! You’ll see the exact test I ran and the two fixes you can copy-paste today to keep your own endpoint off the hacker menu. Catch you in the comments,Diogo
about 1 month ago • 1 min read
Hey Reader, Apple just rolled out a deep dive that claims our smartest chatbots hit a brick wall when puzzles get tough. The paper—titled “The Illusion of Thinking”—fans out six large models and asks if slow “thinking” text actually helps them solve logic games. I read the whole thing so you don’t have to, and here’s the story in plain talk. HOW THE EXPERIMENT WAS SETUP Apple picks puzzles with clear rules like Tower of Hanoi, river crossings, the 24 Game, and simple path-finding mazes. They...
3 months ago • 3 min read
Hey Hey! Berlin’s in full-tilt summer mode—street raves, open-air cinemas, midnight kebabs. I kept pace and shipped a MASSIVE stack of course upgrades. Jump in, then jump outside. Ninja Haircut GIF RAG, AI Agents & Generative AI with Python and OpenAI 2025 NEW IN MAY: RAGAS + RAG deep dive with all the latest OpenAI tricks—done and waiting COMING IN JUNE: New sections on OpenAI Image Generation + Reasoning Models AI Agents For All! Build No-Code AI Agents & Master AI 2025 MAY REMAKES:...
3 months ago • 1 min read
Hey Reader! The internet is filled with ChatGPT failures. From the inability to count the R's in the word strawberry: Asking ChatGPT to count the # of Rs in the word strawberry To being unable to answer simple puzzles correctly: Puzzle that ChatGPT cannot answer But the bottomline is that this goes way beyond simple things. Go to ChatGPT and type: Generate an image of a clock And you will get something like this: Image generated with the prompt above Notice the 10:10 time. ChatGPT almost...
4 months ago • 1 min read
Hey Reader! As you can see below, I am permanently back in Berlin. Loads of new stuff added. But, it is May Day. One of the most legendary days in Berlin. And, from my perspective, fun is part of the curriculum or, you know, life. Don't skip on it. Im Ready To Party Kristen Wiig GIF RAG, AI Agents and Generative AI with Python and OpenAI 2025 Sections on Retrieval Fundamentals, Generative Fundamentals and Introdution to RAG had their Python tutorials remade. NEW Sections: LightRAG and OpenAI...
4 months ago • 1 min read
Hey Reader, This has never happened before. Starting today — for this entire week — every single course inside ZTM is 100% free. No credit card. No tricks. No strings. Just create a free account and start learning: 👉 Join Free Week Here Why are we doing this? Simple. We believe once you experience how good ZTM is — the courses, the community, the results — you’ll want to stick around. But this is a one-time experiment. It ends on April 21. So if there’s a course you’ve been eyeing... If...
5 months ago • 1 min read
Hey there, Do you want to crush it in the AI space? I’ve just released a brand-new No-code AI Agents Course on Udemy, and I want you to see the promo video. This course shows you how to master AI Agents so you can stay ahead of the crowd. Plus, I just dropped a YouTube review of Kimi AI K1.5, the new AI model from Moonshot AI. I compare it to DeepSeek R1 and even OpenAI’s models. You’ll see how to use Kimi AI, and find out if it really beats DeepSeek and OpenAI. Let me know how you like them....
6 months ago • 1 min read
Reader, Picture this. You’re juggling emails. Slack messages And half a dozen tasks. And it’s only 9:00 AM. You’re tired. You’re stressed. And you’re tired of being stressed. Listen. It's time that we have the talk. I just launched AI Agents For All! Build No-Code AI Agents And Master AI This isn’t your run-of-the-mill “Hey, let’s talk about GPT for fun” course. This is your rocket-fueled fast pass out of grunt work and guesswork. Stop wishing and start building. I’ll show you how to create...
6 months ago • 1 min read
Hello Reader, Hey [Recipient Name], I came across OpenAI’s latest approach for managing AI models, called Chain-of-Thought (CoT) monitoring. It’s all about transparency—letting AI models openly share their reasoning steps so that we can catch and address misbehavior quickly. In practical terms, this means models explicitly tell us their reasoning behind every decision, revealing patterns we normally wouldn't see. For instance, the first screenshot below shows how CoT monitoring encourages AI...
6 months ago • 1 min read
Hey Reader, For years, autoregressive models (ARMs) have dominated the landscape of large language models (LLMs). They generate text one token at a time, a process that has proven effective but comes with inherent limitations. What if there was another way—one that could model entire sequences holistically rather than predicting words step by step? Enter Large Language Diffusion Models (LLaDA), an approach inspired by the success of diffusion models in image generation. Instead of relying on...
6 months ago • 1 min read