We're one week into the AI Second Brain cohort, with two to go.

Participants spent the first sprint building their Master Prompt and putting the PARA skill to work with Claude on their own files.

The Circle discussion space has been full of before-and-afters.

My favorite came from Randa: "The Downloads folder after running the PARA skill brought tears to my eyes!"

Others are reporting hours saved, chaotic folders made sense of, and personal insights from the Master Prompt process itself.

What's struck me most is how quickly people are moving from "learning about AI" to using it on the corners of their digital lives they've neglected for years.

I'll share more as the cohort progresses.

Want to participate in the next one in the fall? Join the waitlist here.​

The 4 AI Decisions You Should Make

Most AI frustration I see isn't about the tool but about the setup around it.

In this video, I walk you through the four decisions that shape how much you get out of AI:

• which platform to use
• which harness to run it in
• where to interact with it
• which model to pick for which task

Get these right, and you'll notice the difference immediately.

Stop Building the System. Start Using It.

One of the most common AI mistakes I see — and one I've made myself — is trying to build everything before doing anything.

In our AI Second Brain cohort, I call it "boiling the ocean." You want a system that works for every project, every tool, every scenario. So you spend hours setting it up. And you never actually use it.

Here's what works better:

• Pick one task you need to do today
• Gather only the files and notes relevant to that task
• Hand that context to your AI and get to work

You don't need a complete system. You need enough context to solve the problem in front of you — right now.

Your action step: What's one task on your plate today? What's the minimum context your AI would need to help with it? Try it, then reply and tell me what happened.

The Throughput of Learning

Here's a piece I wrote back in 2017 that I keep coming back to.

The argument: learning isn't about collecting more information. It's about how quickly you can invalidate the assumptions you've been operating under.

Every stage of life shifts the bottleneck somewhere new. Access to information. Then structure. Then synthesis. Then insight.

If you've been saving content all year and wondering why it doesn't add up to much, this one might reframe the problem for you.