The 7 habits that make AI simple

I was quietly convinced I'd never get good at AI

June 02, 2026

For a long time I did the thing most people do. I opened a chatbot, typed a vague question, got a mediocre answer, and quietly decided AI just wasn't for me. I nodded along with everyone else who said the same thing.

Then I came across a breakdown that reframed the whole thing for me. The argument was simple and it stuck: mastering AI isn't about being technical. It's about a handful of repeatable habits. The creator laid out 7 skills that turn the whole thing from intimidating to almost embarrassingly simple, and I was impressed enough by how practical each one is that I broke them down here.

I read it twice. Then I started running the habits myself. Here's what each one actually means.

In partnership with

AI agents now read your docs almost as much as humans do.

Mintlify analyzed 790 million requests across its documentation platform. The finding: AI coding agents account for 45.3% of all traffic, nearly tied with traditional browsers at 45.8%.

Two tools are driving almost all of it:

  • Claude Code: 25.2% of total traffic, more requests than Chrome on Windows

  • Cursor: 18% of total traffic

  • Together they account for 95.6% of all identified AI agent traffic

The rest of the field, OpenCode, Trae, ChatGPT, and NotebookLM, is showing up but nowhere close.

One caveat: OpenAI's Codex doesn't send an identifiable user-agent header, so the real agent percentage is likely even higher.

The takeaway for anyone maintaining developer docs: your documentation now serves two audiences. Structure and machine-readability matter as much as clarity for human readers.

*Ad

The 7 skills that do the heavy lifting

  1. Stay updated, but barely. You don't need 40 newsletters and a doomscroll habit. Pick 2-3 creators who actually teach AI step by step. One newsletter, once a week, done. The real trick is what comes next: every article you read, try one thing immediately. Not later. Now. Reading without doing is how knowledge evaporates.

  2. Pick one tool and master it. Here's where most people go wrong. They collect tools like trophies and master none. The fix is brutal and simple: pick one, delete the rest from your bookmarks, and use only that tool for 30 days. Go deep. Learn its projects, its memory, its search, its file uploads. Depth beats variety every single time.

  3. Set up your AI before you prompt. Most people prompt cold, with zero context, then blame the AI for generic output. Instead, create a folder called "AI Files." Your first file describes who you are, your tone, your audience. Then the workflow becomes: upload files, define the task, define what success looks like. Setup is the unfair advantage almost nobody bothers with.

  4. Teach AI what you know. This is the cleverest one. Flip the script and let the AI interview you. The prompt is simple: "Ask me questions about my expertise." Let it pull out your rules, your hard no's, your audience. Then export all of it into one .md file you can reuse for months. You're basically cloning your own brain into a reusable brief.

  5. Talk to AI like a colleague. Treat it as a thinking partner, not a vending machine. Start with "Don't start yet. Ask me questions." Read version one, then name everything that's wrong with it out loud. Then push harder with a line I now use constantly: "Argue against this." Friction makes the output sharper.

  6. Ship before it's perfect. This one stings a little because it's true. Build the rough draft with AI in 20 minutes. Show it to real people and let them react to something tangible. You sell what it becomes once others invest in it, not the polished thing you hid for three weeks.

  7. Lead AI, don't follow it. Stay in the driver's seat. Split every task into two questions: what does AI do, and what do I do? Hand it the 80 percent of grunt work. Keep the 20 percent that needs your judgment. And the line that stuck with me most: if you can't spot the mistake, don't delegate it. You can't lead what you can't check.

How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

*Ad

Why these beat any single prompt trick

What I appreciate about the framing is that none of these require a computer science degree. They're behavioral. Pick less, go deeper, set context first, treat the AI like a sharp colleague, and stay the one making the calls.

Strung together, they compound. Most people chase the perfect prompt, the magic phrase that unlocks everything. There isn't one. There's just the boring stack of habits that makes every prompt you write land better than the last.

The line about 2030 I keep coming back to

The breakdown also makes a bigger prediction worth sitting with. Here's how the creator put it:

The skills that matter in 2030 are the ones AI makes 10x more powerful, or the ones it can't touch at all. The middle gets automated.

I keep coming back to that middle line. The safe, average, in-between work is exactly what gets swallowed first. The way to stay valuable is to either become someone AI amplifies dramatically, or to own the judgment and taste that AI simply can't replicate. These 7 habits are how you position yourself on the right side of that split.

Where to start tonight

If you only act on one of these, the advice is clear: pick a single tool and commit to it for 30 days. Then layer in the context setup so your AI actually knows who it's working for. Small, boring, repeatable. That's the whole secret.

My honest take

I spent a long time treating AI like a slot machine, pulling the lever and hoping for a good answer. The shift that mattered wasn't a better prompt. It was deciding to go deep on one tool and feed it real context before asking for anything.

The habits that worked for me are the unsexy ones. One tool. One context file. One question at a time, with me still holding the wheel.

So pick your one tool right now, open a blank file called "AI Files," and write three sentences about who you are and who you're writing for. That's the whole first step. Do it tonight, not later.

Hiring in 8 countries shouldn't require 8 different processes

This guide from Deel breaks down how to build one global hiring system. You’ll learn about assessment frameworks that scale, how to do headcount planning across regions, and even intake processes that work everywhere. As HR pros know, hiring in one country is hard enough. So let this free global hiring guide give you the tools you need to avoid global hiring headaches.

*Ad