This Is Not a Blog Post About Hair Care. It's a Blog Post About AI.

Though if you only skim, you might get confused.

You’re reading Playbooks & Priorities, a newsletter about working, parenting, and working parenthood.

---

I’ve been following Abbey Yung for a while now. She’s a social media content creator and trichologist who helps people repair damaged hair and grow it long and healthy. I’ve never been someone who loves hair care or makeup. Anyone who knows me in person would call me low maintenance, as close to no-maintenance as possible.

But I’d also be lying if I said that when my hair started falling out after having my first baby, it didn’t bother me. Since then, I’ve had two more, and that postpartum shedding period is both no joke and no less a hit to one’s confidence. Now that I’m done having kids, I was looking to get my hair back to its low-maintenance state. That’s when I stumbled on the world of Abbey Yung.

It wasn’t just watching her videos that got me interested- it was watching the reviews, the other people talking about their effects. I decided to give her system a go. But honestly? I found it overwhelming.

The Abbey Yung Method has 11 steps with a bunch of substeps. All in, it’s almost twenty decisions! Let me break down what this actually looks like:

Pre-shower, you might do a bond repair treatment or an oil treatment - but only sometimes, not every time. In the shower, you need to pick the right shampoo for today (clarifying at least once a week, but also maybe non-clarifying or medicated depending on what’s going on with your scalp). Then there’s bond repair treatment, but the frequency varies by which specific treatment you’re using. Then conditioning - but you’re supposed to pick 1-2 products from options that include glosses, conditioners, and masks.

Post-shower, there’s potentially another bond repair treatment, then leave-in conditioner and heat protectant every single time, plus optional styling products and style sealers (serums, lotions, oils). And between washes? More optional treatments, more heat protection if you’re styling, and dry shampoo.

Each of these steps has multiple product options. Some steps need to be done every wash, some once per week, and some are optional entirely. And you’re supposed to remember all of this while standing in your shower with wet hair.

Who wouldn’t feel overwhelmed trying to keep track of all that?

So, I started with the first one: the Epres Bond Repair spray. I bought it online, had it shipped to my house, and was flabbergasted at how expensive it was for that little bit of concentrate. I decided to give it a go. That’s all I changed: I started by completely covering my hair with the spray using their bottle and following some of the example videos I found online. I did that once a week for a couple of weeks, and it made a huge difference. It wasn’t overnight- it wasn’t that same wash day- but I did notice a difference, and the only thing I had changed was that one spray.

That’s where I decided to double down on the AYM (Abbey Yung Method). But I was still feeling a bit overwhelmed by all the steps.

So I used AI to solve this problem.

Here’s What I Did

First, I found the canonical Instagram post explaining the Abbey Yung Method. I took screenshots of the whole thing and fed them into Claude to turn it into written explainers. No more scrolling back through Instagram trying to remember if I was supposed to use the bond repair before or after conditioner, or whether today was supposed to be a clarifying day. (On a personal note, I don’t have an Instagram account, and it is nearly impossible to use Instagram without an account.)

Second, I used that explainer to create a product spec. I picked one product for each category and simplified the whole system. I didn’t need to see every possible option - I just needed to know what I was going to use and when. I didn’t worry about optimizing for my hair's porosity (who knows?) or texture (it depends on the day of the week). I know I like to wear my hair down, only need to wash it 1-2 times per week, and it holds a ponytail mark pretty aggressively. I simplified the method in a way that worked for me.

Third, I put my product spec into Kilo Code’s App Builder. After about an hour of back-and-forth and guidance, I had a custom web app that helped me navigate the method. It helped me know which products I needed in what order, which steps were optional, and how often I wanted to consider using each of them.

That was in December. I’ve been using it for two months now, and I’ve kept improving it. And by “I,” I mean I’ve used Kilo Code and AI to iterate on it. I haven’t written a single line of code myself.

What Iteration Actually Looks Like

In the first version of the app, I was focused on guidance - making sure I knew what to do in the right order. The app would walk me through the steps: clarifying shampoo today, then bond repair, then this conditioner, then leave-in and heat protectant. It was basically a checklist that helped me navigate the 20-step method without getting lost.

And you know what? That first version was useful. I could actually follow the Abbey Yung Method without constantly pulling up Instagram or second-guessing myself. But as I kept using it, I realized I had a different problem I hadn’t anticipated: I needed to remember what I’d done last time.

I’d stand in the shower thinking, “Wait, when did I last use the glossing mask? Was that last week's hair wash or two weeks ago? Do I need to clarify today, or can I skip it?” The app told me what to do this time, but it didn’t remember what I’d already done.

I couldn’t have known to build tracking into the first version. I didn’t understand my real problem until I was actually using the tool. This is how software projects actually work - you build something, you use it, you discover what’s missing, you build that next.

This is also where AI changed the game for me. In the past, going back to refine a tool I’d built would have felt like a big deal. I would have needed to remember how I built it, where the code was, and what I was thinking at the time. But with AI, iteration is easy. I just opened up the app, told Kilo Code, “I need to track sessions - when I last did each step,” and it updated the app.

So I added session tracking. Now every time I wash my hair, I log it. The app remembers when I last did a clarifying wash, when I last used the bond repair, and when I last did a glossing mask. It can tell me “you used the Epres on Monday, you’re good to skip it today” or “it’s been 10 days since you clarified, you should do that.”

Then I made another change. When I first built the app, I was tracking the abstract steps from Abbey’s method. But I kept having a hard time tying steps to specific products - her method lists multiple options for each step, and I’d forget which products I’d actually decided to use.

Here’s the thing: I wasn’t building this app for some general use case. I wasn’t trying to create a tool that anyone following the Abbey Yung Method could use. I was building it for me. So I did what you should always do when building for yourself - I added the specific products I actually use to the app.

Now, when it’s time for a bond repair step, the app doesn’t show me three options and make me remember which one I bought. It shows me: “Use Garnier Fructis Hair Filler Bonding.” When it’s clarifying wash day, it tells me: “Use the L'Oréal Ever Pur.” When it’s time for leave-in and heat protectant, it doesn’t list serums, lotions, and oils - it just tells me which one I’ve chosen to use: OGX Bond Repair Protein.

Specificity is a feature, not a bug. When you’re solving your own problem, you don’t need flexibility. You don’t need to account for edge cases. You just need it to work for your specific situation. And that’s way easier to build.

As I’ve made these changes, I’ve also gotten better at prompting. I understand better how to describe what I want, how to break down a feature request, when to be specific, and when to let the AI figure out the details. But here’s the thing: I learned that by doing it, not by reading prompting guides or taking courses. The best way to learn how to work with AI is to have a real problem you’re trying to solve and iterate your way to a solution.

The app I have today is different from the one I built in December, and it’ll probably be different again next month. That’s what iteration looks like. Not some grand redesign. Just continuous small improvements based on actually using the thing.

This Is How You Should Be Using AI

Not for generating blog posts you could write yourself. Not for summarizing meetings you should probably have paid attention to. Not for writing emails that sound like everyone else’s AI-generated emails.

Use it to solve real problems in your life. Problems that are too small or too specific for anyone to have built a solution for, but annoying enough that they’re actually getting in your way.

The Abbey Yung Method is genuinely good for your hair. But 20 steps is genuinely too much to track in your head. I tried using a spreadsheet first - but honestly, building the app was easier. That’s where we are now. For many real-world problems, it’s easier to build a custom tool than it is to wrangle a spreadsheet.

This is AI literacy. Not knowing how to prompt. Not understanding transformers or tokens or whatever technical concept is trending this week. AI literacy is:

1. Recognizing when you have a problem that could be solved with software. Standing in the shower, trying to remember 20 steps? That’s a software problem, not a memory problem.

2. Breaking that problem down into pieces. Screenshots of Abbey’s Instagram → Claude turning them into an explainer → a simplified product spec → a working app. Each step is manageable. The whole thing together is magic.

3. Using AI tools to build the solution. One hour with Kilo Code’s App Builder. No coding required. Just describing what I needed and refining until it worked.

4. Iterating on it as you actually use it and find the rough edges. The first version was a checklist. Then I needed tracking. Then I needed my specific products. Each version got better because I was actually using it.

The skills you need are the same skills you’ve always needed: critical thinking, problem decomposition, the ability to spot when something isn’t working, and figure out why. The difference is that now you can actually build the solution yourself instead of either living with the problem or hoping someone else builds it for you.

So What Happened to My Hair?

It’s doing so much better! Turns out following a 20-step method is way easier when you have an app telling you exactly what to do and tracking everything for you. Who knew?

But more importantly, I built something useful in an hour. I’ve continued to improve it over the last couple of months.

That’s the future of work. Not AI replacing you. AI letting you solve problems you couldn’t solve before.

So what’s your Abbey Yung Method? What annoying, specific problem is sitting in your life right now that you’ve been working around instead of solving? Open Kilo, describe one annoying thing in your life, and ask what it would take to solve it. That’s where you start.

Link to the app if you want to see it in action.