It's gotten to a point now where I need to run two machines. This has gotten to a new level already. Just 2 days after @Igor Pogany gave us a behind the scenes run down on how we can really utilize things even more. I've had to use my main machine as my M365 Copilot system while having another machine that runs my Perplexity, ChatGPT, Claude, and others. I have Claude working overtime right now going through EVERYTHING on my enterprise system in Microsoft with my Agent that I duplicated from Copilot to Claude. Claude is my knowledge center while Copilot is my fully integrated multi agent system. Perplexity is the research engine that that ChatGPT uses for research. ChatGPT builds the framework from there and then sends it to Claude for Claude to use with my Project agent Atlas that is the Mirror to my Copilot Agent Atlas that has the 8 sub agents. This game is WILD!!!!

After today's live, I'm sitting at 11 agents in M365 Copilot, plus adjacent Project Agents running in Claude, Perplexity, and ChatGPT. Here's how it breaks down using the 3 agents I actively interact with: 1 Business agent → orchestrates 8 sub-agents underneath it 2 Personal agents: - Agent 1 = Brain/big picture: long-term decisions and outcomes - Agent 2 = Thoughts/short term: daily/weekly/monthly execution, research, and task orientation. Also acts as a backup/support layer for Agent 1. The structure is what's making this work. One layer thinking long-term, one layer executing short-term, and a business layer with its own delegated sub-agents underneath. Curious how many of you are running multi-agent setups like this vs. one agent doing everything? What's your split?

Which LLM's are best for having speak to one another? Right now I'm using Claude and ChatGPT to speak back and forth to build me an Asphalt Consulting Firm as that's what the majority of my experience is in. Let me know what works best for you all?

@Igor Pogany Give us some insight into what we are to expect on Saturday? You've already given us some great stuff to work with. Is this a chance for you to show us a little behind the scenes of how you and Dean make it all work? Either way, it's going to be great stuff as always! Show us more about this smoke machine!!

I'd love to hear from people who have moved beyond experimenting with AI and are now using AI agents in their day-to-day workflow. I'm particularly interested in questions like: - Which parts of your writing business have you automated? - Which AI tools or platforms are you using (Claude, ChatGPT, n8n, Make, etc.)? - How much time does it genuinely save you each week? - Which tasks still require a human touch? - If you were starting again, what would you build differently? Whether you're an author, publisher, content creator, or run another writing-related business, I'd really appreciate hearing about your real-world experiences. I'm trying to understand what's actually working in practice, rather than what's theoretically possible. 😊