Hey friends,

Happy Wednesday morning from sunny Northern California.

I have to say it. I am a huge fan of daylight savings time. I know that is a controversial take. But waking up to this spring weather with the extra light in the evening? It hits different. Especially when you are about to have one of the most productive mornings of your life.

Let me tell you about this morning.

I opened a tool I had never used before. It's called Cowork. Anthropic's new desktop agent, still in research preview. I'd been meaning to try it. I didn't have a plan. I just opened it and started talking.

Four hours later I had mapped three years of my own work. I had audited all three of my websites. I had a clear revenue strategy, a product offering with pricing, a tech stack that finally made sense, and a launch plan for the founding member tier I'm about to tell you about.

Dictated. Not typed.

No switching between Claude and ChatGPT and Perplexity and Gemini. No losing my train of thought every time I opened a new tab. No re-explaining who I am and what I'm building to a tool that forgot me the moment I closed it.

One conversation. Four hours. Three years of work, finally connected.

And here's the thing that stopped me cold somewhere around hour two.

I counted my tools. I had 25. ChatGPT, Claude, Gemini, Perplexity, Notion, Fireflies, Otter, Whisper Flow, Grammarly, GoHighLevel, Canva, Loom, Beehiiv, Bolt.new, Firebase, GitHub, NotebookLM, Copilot. I'm not going to list all of them. You get the picture.

Twenty five tools. Not one of them knows who I am when I open it.

Every single day I spend a piece of my brain budget re-explaining myself. To the AI that forgot me overnight. To the notes app that has no idea what I was thinking last Tuesday. To the research tool that treats me like a stranger every single time.

For a dyslexic brain, that isn't just annoying. It's a tax. A real cognitive tax that compounds every single day. And I've been paying it for three years.

I built the mess so you don't have to.

What Anthropic Just Built Changes Everything

While I was in that four hour session this morning, I learned about two things Anthropic released that I think are the most relevant tools our community has seen yet.

The first is the Memory Tool. It's an API-level feature that lets Claude create, read, and update files in a persistent memory directory. Those memories survive across every conversation. Claude checks its memory before starting any task. Your cognitive profile, your projects, your preferences. all of it stays, grows, and travels with you. Client-side, which means your data stays under your control. Sovereignty by design.

That is the Personal Knowledge Graph I wrote about in the Cognitive Partner Model whitepaper. Anthropic just built the infrastructure version of it.

The second is Import Memory. available right now at claude.com/import-memory. You go to your other AI tools. ChatGPT, Gemini, wherever you've been working. run a specific prompt that pulls out everything those tools have learned about you, copy the output, and paste it into Claude's memory settings. Under a minute. Your context travels with you. You stop starting from zero.

I know. I know. We've been talking about this exact thing for as long as I've been writing this newsletter. AI that knows you. Context that persists. A cognitive partner, not a chatbot.

It's here. Right now.

But Here's What I Really Want to Talk About

In Edition 329 I showed you how to build a Single Source of Truth. a document that tells your AI who you are so it stops hallucinating about you. We talked about mise en place. The filing cabinet versus the workbench.

This morning I realized something. Most of the people reading this haven't been doing this for three years across 25 tools. Most of you are at the beginning. Or you've tried a few things and it hasn't clicked yet. Or you open ChatGPT, stare at the blank screen, don't know what to ask, and close the tab.

I built the mess first and learned the hard way. You don't have to do that.

What I want to give you is the clean path. The "start right" version. The system you set up once that makes every AI you open actually useful from day one. because it already knows you.

In Edition 329 I called it Building Your Second Brain. This is the live version. The one we build together.

Three Things That Happened This Morning That I Keep Thinking About

The first is that I used a tool I'd never tried before and it was immediately useful. No tutorial. No onboarding. I just started talking. The way I think, which is sideways and fast and out of order. And it kept up. That's what a cognitive partner feels like. That's what we've been describing for 332 editions. And today it was just... there.

The second is that I realized my mess is the map. Three years of scattered conversations across six AI platforms, 148 files in Google Drive, three websites with no clear funnel between them. that chaos is actually a map of exactly what not to do. And maps are valuable. You can hand someone a map.

The third is that I realized I've already built most of what I've been writing about. The Cognitive Partner Model. The 10-80-10 Rule. The Single Source of Truth. The research pipeline. The voice-first workflow. I've been living it and writing about it. I just haven't packaged it for you yet.

Today that changes.

What I'm Building. And Who It's For

I'm launching the Cognitive Partner Membership.

This is for you if you've been reading this newsletter and thinking "I want that. I want to feel what Matt describes when he talks about flow and cognitive partnership and AI that works the way his brain works. but I don't know where to start."

It's for you if you're neurodivergent and you're tired of AI tools that feel like they were built for someone else's brain.

It's for you if you've tried Claude or ChatGPT a few times, liked what you saw, and want someone to help you set it up properly from the beginning. so you don't spend three years building the mess I built.

It's also for you if you ARE me. three years deep, tools everywhere, wanting to finally pull it all into one place.

Here's what's inside:

The Dyslexic AI Assistant. a pre-configured AI hub built on the Cognitive Partner Model. Voice-first. Phonetic-spelling friendly. Pre-loaded with the CPM prompt library. Multiple AI models, one interface. You open it and it works. You don't choose settings. You don't write prompts from scratch. It's ready for how your brain works.

The Cognitive Partner Setup. a 30-day onboarding sequence that walks you through the exact steps I took this morning. Week one: import your context from every AI you've ever used. Week two: build your Single Source of Truth. Week three: set up your research digest so the news you need comes to you. Week four: run your first real project through the 10-80-10 workflow.

Monthly Live Workshop. one hour, my real screen, showing how I actually use the tools. Not slides. Not a course. My Tuesday morning, live. Recorded for members who can't attend.

The Research Digest. curated weekly roundup of AI and neurodiversity news, filtered through CPM parameters. I'm consuming this content anyway. Now it comes to you instead of staying inside my head.

What's Coming This Week

This was the biggest week in AI this year. And I am not going to try to cram it all into one edition.

So this week you are getting a three-part series.

Edition 334 covers the Anthropic labor market study that dropped the same day as Import Memory. Real data on which jobs AI is actually affecting. Which occupations are most exposed. Why the window to position yourself is closing. And why that gap between what AI can do and what people are actually using it for. the same gap I just described with my 25 tools. is your biggest opportunity right now.

Edition 335 goes deeper on cognitive partnership. Now that the Memory Tool and Import Memory exist, what does it actually mean to build a cognitive profile? How does it connect to the Cognitive Balance Model? And why is this especially transformative for neurodivergent thinkers?

Edition 336 covers OpenAI's GPT-5.4 release. A million token context window. ChatGPT for Excel. And what the pace of AI acceleration means for everything we have been building.

Three articles. One week. All connected.

Because this stuff does not exist in isolation. The tools, the data, and the frameworks all point in the same direction. And that direction is exactly where we have been heading for 332 editions.

Here's what I know after this morning.

The cognitive partner isn't a fantasy. It's not a whitepaper idea or a newsletter thesis. It's a Wednesday morning in March 2026, spring weather, extra daylight, using a tool you've never tried, having the most productive strategy session of your life, and realizing that three years of work you thought was scattered is actually a foundation.

Your brain was never the problem.

The interface was. And now we're fixing it.

Let's build.

Dictated, not typed.

🧠 The Story: I spent four hours this morning in one AI conversation and mapped three years of my own business. Dictated, not typed. On a beautiful spring Monday in Northern California.

⚡ The News: Anthropic just released the Memory Tool and Import Memory. you can now bring your ChatGPT and Gemini context into Claude in under a minute.

🔧 The Insight: You don't have to build the mess first. There's a clean path. I'm handing it to you.

🚀 The Announcement: The Cognitive Partner Membership is open. 50 founding member spots at $19/month locked forever.

📰 Coming This Week: A three-part series on the biggest week in AI this year. Edition 334: Anthropic's labor market study and your career. Edition 335: Cognitive partnership and your brain's profile. Edition 336: GPT-5.4 and the acceleration of AI.

⬅️ Previous: Edition 332. Meet in the Middle | Edition 329. Building Your Second Brain ➡️ Next: Edition 334. The Data Is In. Here's What AI Is Actually Doing to Jobs Right Now.