I was supposed to be doing something productive. Instead I watched Mark Kashef's video on combining Claude Code agent teams with Nano Banana and thought "I could set that up right now." as I seem to have a habit of doing! 15 minutes later, I had a working agent team generating scored image variants for a workplace mental health graphic. The screenshot below shows the critic agent's ranked output. For anyone unfamiliar, the approach comes from a Google research paper called Paper Banana. The core idea is treating image generation like a design agency: a research agent analyses reference images, a prompt architect creates narrative prompts, a generator calls the Gemini API, and a critic agent scores the results on faithfulness, readability, conciseness, and visual appeal. Then it iterates. What impressed me was the critique loop. Instead of generating one image and hoping for the best, the system produces multiple variants and ranks them against defined criteria. That is genuinely useful for someone without design skills who still needs professional-looking materials. It needs fine tuning for my specific use case, but the foundation is solid. I find with these things that if I do not set them up while I have the headspace, they never get done. So down the rabbit hole I went. Full credit to Mark Kashef for the walkthrough: https://www.youtube.com/watch?v=mGfQV4s1MgE Ryan has had Mark on his channel before so I suspect this will not be entirely unfamiliar territory. But if you have not seen agent teams used for image generation with a built-in feedback loop, it is worth 14 minutes of your time.

Would love to hear yours! I‘ll start: - em dashes (I mean come on, still?) - overly polished - emotionally balanced - very well structured (kind of a good thing, right?) - "it’s not just X, it’s Y" - soulless (hard to say why but you just feel it)

Hi Ryan! With your recent move away from ChatGPT and to Claude/Google, I was wondering if you have any updated tips on how to use AI to find YouTube video ideas within my niche (songwriting). While I have a ton of ideas for content, I'm specifically trying to identify "outlier" topics that are more likely to get traction/engagement. Any thoughts on this? Thank you!

People talk about AI saving time, but it can be hard to picture what that means day to day. Here is what one day actually looked like for me. Using Claude Code through VS Code, and in between marking course assessments, I have: Built a new campaign landing page from scratch. Then created a full campaign outline with step-by-step tasks, some of which Claude can execute directly with my oversight. Restructured my Airtable CRM. Not just tidying fields. Claude suggested better ways to connect contacts, companies, enquiries, and events together, then built the formulas, rollups, and relationships to make it work. This is the kind of structural thinking that would normally need a consultant or developer. Built a full automation logging and monitoring system across my n8n workflows. Every workflow now logs success or failure to Airtable automatically, with an error catcher and a daily summary email. I went from hoping my automations were running to actually knowing. None of this was copy-paste prompting. It was collaborative. Claude suggested approaches, I pushed back or redirected, and we iterated. Some things took multiple attempts. But the volume of meaningful work completed in a single day would have taken me weeks working alone or cost thousands in consultant time. For anyone still thinking about trying Claude Code: it is not a chatbot that writes blog posts. It is a working partner that builds real systems. The more I keep using it, the better it gets.

Big week for model releases (again). So I updated my app. Check out the latest and greatest.🤖💥 Hey Ry - any chance you doing a tutorial with the mistral/Voxtral models... Voxtral looks amazing.