Hello everyone! I'm Kevin and I've just started the 7-Day Challenge. In the short term, I will build an AIOS that will help me with my marketing goals for my event production company and my lady's personal training brand. In the long term, I will create a full-fledged AI Agency with services in not only just marketing but branding, content creation, webdev/SEO, operations, sales, logistics, and anywhere else from the tech-facing side that would find immediate benefit using automation tools and custom-built software to fulfill these promises. And even further down the line, I'm going to school for electrical engineering and robotics right now, so I plan to utilize what I learn in building this AIOS agency/project to integrate my systems into the physical world too. My experience level is pretty average. I've mostly used Gemini, ChatGPT, and Claude almost entirely as just chatbots over the last year or so, and recently started using Claude Cowork (although I don't like it by itself for building an AIOS) but I'm in the process of switching over to Claude Code as I prefer the control I can have over it. What brought me here was coming across Nate's Youtube videos multiple times in my Youtube and Facebook feed and realizing I really like his educational content and tutorials. His use of Claude Code and n8n has really peaked my interest, so I've finally decided to take the dive and jump into it. Thanks for having me everyone, I'm pumped to be here! I'm not an expert by any means, but I like to help out whenever I can so I'm sure you all will see me around. #AISChallenge

I'm in the middle of creating the newsletter automation workflow for the initial project and Claude randomly decided to name the plan's md file as i-want-to-build-shiny-cook.md which made absolutely no sense to me. So I decided to ask it why, and I've attached its response. For reference, I'm running Sonnet 4.7 Medium with Adaptive Thinking turned off. I copy and pasted the exact prompts Nate used in his video so I have no clue where it got either of those last 2 words from. But, I'm gonna keep it named that for now because I find it really funny and it's just a demo project anyways. Has anyone else ever seen this kind of odd behavior before? Any ideas as to why it's happening?

I've been building the architecture for an AI Agency with the premise of utilizing Claude Cowork on and off for the past couple weeks now. Finally finished my "alpha" v1.0 and immediately started running into problems that I didn't think would be prominent until later on as the work got more complex. First off, building parent-child relationships between Cowork projects is a bad idea lol. I have an Agency Central project as the main parent with individual Client projects as linked children. However, since instructions are native to the UI and not inherited naturally through the file architecture, this makes instructions communication difficult to say the least. One option is to only use the native UI project instructions for super minimal basic instructions across all projects, and then still write independent agency-project-instructions.md and client-project-instructions.md within the file architecture. However, this now requires a sync protocol to be established so that the agency project parent instructions are copied across all client project roots (in an About Me folder) which seems like an entirely unnecessary waste of tokens. Not to mention there's like 3 more .md files I've created (agency-profile, anti-ai-writing-style, and tool-router) which also need to be synced across all client projects as well. With Claude Code, instruction inheritance is already incorporated. The other issue I have is with control within Cowork. You can't switch models, adaptive thinking, or effort levels mid-session. It requires a whole new session. Also, any connectors, plugin, or skills are ALL loaded up in every single session no matter whether you use them or not, which absolutely destroys token usage. Not to mention that I haven't even installed any yet and I'm still maxing out my 5hr usage limits on the Pro plan anyways. Seems like these are things Anthropic could easily change in Cowork, but to be honest idk shit about developing so I'm probably talking out of my ass. And if it's natively available in Claude Code, it would seem like I should probably be using that instead anyways.

New here? 👋 Me too...navigating the Skool gate. Since we're both unlocking value: What's the one personal AI automation you've built that you use daily? Any unique workflows that you can't live without? Drop them below🔻

Every single person following AI right now is overwhelmed. Including me. I make videos about this stuff for a living and I still feel the pressure. New model drops. New framework. New feature update. It feels like every single day. But after hearing a ton of you guys bring up "AI overwhelm" week after week, I realized this: → There's a HUGE difference between knowing the "what" and knowing the "how." Staying aware does not mean testing everything. Most new tools and features only need the "what." You see the title. You understand what it does. You move on. The "how" is reserved for the stuff that solves a problem you actually have right now. So when something new drops, I ask myself one question: Does this solve a specific pain point I'm currently dealing with? If yes, I test it in a real scenario. I test it against something that actually matters to me. If no, I save the link. I mentally file it away. And I keep walking. Because here's the thing. Your north star is probably very different from mine. Part of my job is to experiment, form opinions, and share what I think is useful. So naturally I test a lot of stuff. But if your north star is building a business or getting better at your craft, then every shiny new tool might just be a distraction. The number one mistake I see people make is they try to learn everything. They watch every video. They test every tool. They jump to the next thing before the last thing even had a chance to work. And if I've contributed to your overwhelm with my daily uploads, I apologize. hehe. But a lot of people think that this ties directly into how you measure your day. Productivity is not how many hours you worked. It's how many meaningful outputs you created that actually moved the needle towards your north star. Someone can work 12 hours one day and feel insanely productive, but they were just watching tutorials and playing around with new tools. Meanwhile someone else sits down for 5 hours, ships the one thing that actually matters, and makes more progress.