Many people use AI the wrong way.
They open it, type a question, read the answer, close the tab, and that's it. It's a slightly smarter search engine to them. Useful sometimes. Easy to forget.
I was that person. Then something changed, not because I learned a fancy technique or bought a course from somewhere, but because I started bringing Claude into the messy middle of my actual work. Not just the finished questions, but the half-formed ones. The overwhelmed ones. The "I don't even know where to start" ones.
That is when it became a second brain to me.
This is a full breakdown of how I use it. Not a theoretical framework but the real stuff. The workflows, the approach, the honest quirks of actually making this work.
First: What Does "Second Brain" Even Mean?
The original idea of a second brain, made popular by Tiago Forte, is about building an external system to capture, organize, and retrieve your thinking so your head doesn't have to hold everything at once.
The problem with most second brain systems is that they require a lot of upkeep. Folders. Tags. The discipline to file things correctly. Most people set them up and slowly abandon them.
What I found with Claude is something different. It doesn't need to be maintained like a filing system. It's more like a thinking partner that shows up whenever I need it, and it can receive my chaos and give back clarity.
Here's exactly how that works for me.
How I Use Claude as My Second Brain
1. I Pour Out My Thoughts, and Claude Organizes Them
This is the thing I use most and the thing most people should try.
When I have an idea forming in my head, be it for an article, a project, or a strategy, I don't try to write it cleanly. I just open Claude and dump everything. Stream of consciousness. Contradictions included. Half-sentences welcome.
Something like:
"I want to write about why people get overwhelmed with AI tools and what they should do instead, something about how the problem isn't the tools it's how they approach it, maybe also something about starting small, I keep thinking about how my friend asked me about ChatGPT and didn't know where to begin...blah blah blah"
What comes back is organized. The core argument identified. The structure suggested. The noise separated from the signal.
This isn't Claude writing for me but rather it helping me see what I am actually trying to say. That is a different thing, and it is genuinely useful.
Try this. Next time you have a messy idea, don't try to write it out cleanly first. Just talk at Claude like you would talk to a friend. Then ask it to organize what you said and find the main point.
2. I Research Wider, in Less Time
Researching a topic most times means opening fifteen tabs, getting distracted by eight of them, and spending two hours to get a picture that is still incomplete by the time you reach the thirteenth tab.
Now I use Claude to survey a topic first.
I will describe what I am trying to understand, what I already know, what angles I am uncertain about, and ask it to help me map the landscape. It is not a replacement for deep reading (there is no escape from deep reading; you must do it), but it's an incredibly fast way to figure out what to read deeply and where the gaps are.
When I am writing about something, I want to make sure I am not missing an obvious angle. When I make a decision, I want to see all the options before narrowing down. Claude helps me do that in minutes, not hours.
The key: Give it context. Don't just ask a surface question. Tell it what you know, what you are trying to figure out, and what matters to you about the answer. The more context you give, the better the output.
3. It Drafts - Then I Start From There
Here is the thing about AI writing that people either love or misunderstand: Claude does not write for you. It drafts with you.
My actual process is quite simple;
- I give Claude the idea, the angle, the rough structure, and a sense of the tone
- It produces a draft
- I read it, sit with it, react to it
- I edit heavily. Adding what is missing and reshaping sentences until it is right
- Sometimes I go back and forth a few rounds
The draft is never the final product. But it's infinitely easier to work from than a blank page. When I am stuck, having something to push against unlocks the actual writing.
This works for articles, emails, project proposals, creative pieces, really anything. The pattern is the same: direction in, draft out, your judgment applied after.
4. Projects - Code, Planning, and Everything Else
I handle a mix of projects. Some are code. Some are content. Sometimes it is a short story I need to get out of my head. Some are just life admin at scale.
For code projects, Claude is genuinely useful at every stage, right from helping me think through the architecture before a line of code is written, debugging when something breaks, explaining why something works the way it does. I am not always technical enough to know the right question to ask, and that is okay.
For other projects, it functions like a very capable project thinking partner. I will describe what I am trying to build, what constraints I am working with, and ask it to help me map out the steps or identify what I might be missing. It has caught gaps in my thinking more than once.
The habit I have developed is that when I am starting anything new, I brief Claude on it the same way I would brief a collaborator. Context, goal, constraints, what I have already tried. That upfront investment pays off in everything that follows.
What Makes This Actually Work
A few things I have had to learn the hard way.
- Give context generously. Start every session fresh. If you want useful output, you have to be willing to provide context at the start. Think of it less like a search engine and more like briefing someone on a project.
- React to drafts, don't just accept them. The first draft is always a starting point, never a verdict. Tell it what you like, what is off, what is missing. The back-and-forth is where the good stuff usually comes out.
- Use it for thinking, not just producing. The most valuable use I have found is not generating output but rather thinking out loud with something that can help me see my own thinking more clearly. That is underrated.
- Don't outsource your judgment. Claude is not always right. As a matter of fact, it can be confidently wrong. Your job is to bring your own knowledge and taste and verify what matters. It is a tool, not an oracle.
A Note on Getting Started
If you are new to this and the whole thing feels overwhelming, here is the simplest possible entry point:
Next time you have something messy in your head, such as an idea, a problem, a thing you need to write, etc, open Claude and describe it out loud, exactly as it is in your head. Messy and unfinished. Then say something like, "Help me organize this and find the main point."
See what comes back. React to it. Edit it.
That is it. That is the start of a second brain.
The rest of the system builds naturally from there as you figure out what actually helps you think and work better. You don't need to build the whole thing at once.
Final Thought
I think the people who get the most out of AI tools are the ones who stop treating them as answer machines and start treating them as thinking partners.
Your second brain is not a place to store finished thoughts. It is a place to develop unfinished ones.
That is what Claude has become for me. Not a shortcut. A space to think out loud and actually get somewhere.
If you want to build something similar for yourself, start messy. Start honest. See what happens.
Comments
Post a Comment