🎧 How to Use Claude Code as a Second Brain

TL;DR: Today we’re releasing a new episode of our podcast AI & I. Dan Shipper sits down with Noah Brier, the cofounder of Alephic, an AI-first strategy consultancy. Watch on X or YouTube, or listen on Spotify or Apple Podcasts. Here’s a link to the episode transcript.

It’s 4:30 p.m, and after a long day glued to your screen, you’ve finally stepped out for a walk. You’ve barely made it down the block when the dreaded ping arrives: The code in an urgent client deliverable needs a last-minute tweak before it can go out.

Most of us trudge back to our desks, slaves to the keyboard, to do what must be done.

Noah Brier wouldn’t. He’d whip out his phone, tap into his Claude Code setup, and make the change on the spot. And if the stroll sparks an idea for another project—say, an essay he’s been noodling on—he can use the same setup to sift through the approximately 1,500 notes stored inside his note-taking tool of choice, Obsidian, in search of useful nuggets.

While Claude Code is popularly seen as a coding assistant (and he does know how to code), one of Brier’s signature uses is turning it into a research and thinking partner inside Obsidian. Brier tells Dan Shipper how he pulls this off on this week’s episode of AI & I.

Brier is the cofounder of Alephic, an AI strategy consultancy that helps marketing organizations solve complex challenges through custom AI systems. He leads BRXND, a conference about AI and marketing, and previously co-founded Percolate, an enterprise content marketing platform that was acquired in 2019. In this episode, Dan and Brier talk about the nuts and bolts of the latter’s Claude Code-Obsidian workflow, why he insists on keeping AI in “thinking mode,” and how he uses agents inside Claude Code as mental sparring partners.

You can check out their full conversation here:

If you want a quick summary, here are some of the themes they touch on:

How Brier lays the groundwork for his research assistant

Brier runs Claude Code inside Obsidian, the note-taking tool that stores all your notes on a local file-system, organized into folders called "vaults." Instead of running Claude Code on any one of these vaults, he points it at the root directory—the top-level folder where everything lives, so that Claude Code has access to all the notes in his Obsidian.

Brier usually kicks off a new idea by telling Claude Code he’s in “thinking mode,” not “writing mode.” It’s his way of slowing the overeager model down, reminding it to hold off on generating text and instead ask him questions that help him clarify his thinking. Once that frame is set, he feeds it context from so it has the raw material to work with.

On the show, Brier takes Dan through his prep work for a talk he’s giving at BRXND, the AI-marketing conference that he runs. For this, Brier gave Claude Code the big points he wants to make and a working title, “Transformers are eating the world,” following the thesis that LLMs are displacing a bunch of specialized code. He also feeds in an anecdote about the Simple Sabotage Field Manual, a relic from World War II, which I was pleasantly surprised to see in a talk about AI and marketing.

Brier then got Claude Code to go to town on his Obsidian, using it to gather relevant bits of information from the vast expanse of his note-taking history. His instruction is along the lines of, “Go look through the rest of my probably 1,500 things in Obsidian and see [if there’s] anything else you can find that might be of value to this talk.” He isn’t expecting the AI to make any “large conceptual leaps” while pulling notes together. For example, while asking it to find more information about the manual from World War II, it just needed to do a simple search for the articles in Brier’s Obsidian about the topic.

Brier is really just saying: “Amongst this set of things, go find all the notes that I've already researched that brought me to be thinking about these things to begin with.”

Shaping ideas before writing them down

Once Claude Code has pulled relevant material from Brier’s Obsidian into the vault he created for the talk, he starts to sift and recombine ideas. The vault is divided into subfolders: “Chats,” where he saves conversations he’s had with other LLMs using the Obsidian web clipper; “Daily Progress,” where the AI reviews his progress over the day and helps him track how his thinking is evolving; and “Research,” a catch-all for articles, PDFs, and other source material he’s been reading.

According to Brier, “There's entirely too much focus on [AI’s] ability to write and not enough focus on its ability to read… I think arguably [that’s] much more useful on a day-to-day basis. We produce artifacts far less frequently than we just think about things.” (A Claude artifact is a tangible output generated by the model—like a draft, summary, diagram, or piece of code—meant to be saved or reused outside the conversation.)

Agents inside Claude Code as on-call thinking partners

Brier also uses a Claude agent to facilitate his thinking. The agent asks him questions, keeps track of them, and maintains a running log of what he’s uncovering in the process.

One of the biggest benefits, he says, is how it helps him pick up where he left off. AI gives him the luxury of being interrupted and then easily resuming by asking it to “catch me up on the last three days of research,” for example. For Brier, “It's pretty amazing also to just be able to revisit deep work like this… where you know you're gonna break your flow… the hardest part is just picking it up again because you're out of it, and to kickstart that process is amazing.”

Taking Claude Code everywhere

To bring this Claude Code-Obsidian setup onto his phone, Brier uses an app called Termius, which connects to a small computer he keeps running in his basement through Tailscale, a tool that creates a private, secure network so only he can access it. His Obsidian notes are stored in a private GitHub repo, which means he can sync the latest version of his vault onto that computer whenever he needs to. Put together, this lets him open Terminus on his phone, connect through Tailscale, and run Claude Code on all his notes—so he can keep researching and organizing ideas wherever he is, just as if he were at his desk.

This setup has completely changed the way he works. That scenario with the urgent client request? It actually happened. Brier was out on a walk, and instead of heading back to his desk he stopped and sat by a pond, made the necessary fix to the code using Claude Code from his mobile phone, and enjoyed the view.

What do you use AI for? Have you found any interesting or surprising use cases? We want to hear from you—and we might even interview you.

Here’s a link to the episode transcript.

Timestamps

Introduction: 00:01:19
How you can do deep work on your phone: 00:04:28
Why Noah thinks Grok has the best voice AI: 00:06:14
The nuts and bolts of Noah’s Claude Code-Obsidian setup: 00:11:39
Using an agent in Claude Code as a “thinking partner”: 00:23:59
Noah’s Thomas’ English Muffin theory of AI: 00:35:07
The white space still left to explore in AI: 00:44:04
How Noah is preparing his kids for AI: 00:50:41
How he brought his Claude Code setup to mobile: 01:01:54

You can check out the episode on X, Spotify, Apple Podcasts, or YouTube. Links are below:

Watch on X
Watch on YouTube
Listen on Spotify (make sure to follow to help us rank!)
Listen on Apple Podcasts

Miss an episode? Catch up on Dan’s recent conversations with founding executive editor of Wired Kevin Kelly, star podcaster Dwarkesh Patel, LinkedIn cofounder Reid Hoffman, ChatPRD founder Claire Vo, economist Tyler Cowen, writer and entrepreneur David Perell, founder and newsletter operator Ben Tossell, and others, and learn how they use AI to think, create, and relate.

If you’re enjoying the podcast, here are a few things I recommend:

Subscribe to Every
Follow Dan on X
Subscribe to Every’s YouTube channel

If you enjoyed Dan’s conversation with Brier, keep an eye out for a Thesis essay we’ll be publishing by him soon, where he’ll write about how AI can help break through corporate bureaucracy. You can also check out some of our most-loved Thesis pieces by coach to Open AI executives Joe Hudson, founder Sari Azout, and Granola cofounder Chris Pedregal.

Rhea Purohit is a contributing writer for Every focused on research-driven storytelling in tech. You can follow her on X at @RheaPurohit1 and on LinkedIn, and Every on X at @every and on LinkedIn.

We build AI tools for readers like you. Write brilliantly with Spiral. Organize files automatically with Sparkle. Deliver yourself from email with Cora.

We also do AI training, adoption, and innovation for companies. Work with us to bring AI into your organization.

Get paid for sharing Every with your friends. Join our referral program.