I built a career coaching system with AI in days. And got paying customers.

--

I had no idea this was even possible until I tried. Here’s how I built it and what I learned.

A few weeks ago, I found myself in a loop. For the tenth time that month, someone reached out asking for job search help. They’d sent 200+ applications. Heard nothing back. They were talented but invisible in a crowded market.

I’ve had this conversation hundreds of times now. Always the same pattern: talented person, broken system, invisible results.

The advice I give hasn’t changed much over the years. Stop sending applications into the void. Start creating conditions where opportunities find you instead. I’ve used this approach for a decade helping myself, and others, land jobs at companies like On Deck, InVision, Meta, Spotify, and Intercom, often without submitting a single application.

But there was a problem. My time is limited. I can’t personally coach everyone who needs help. And honestly, I’m not interested in becoming a full-time career coach. Plus, it would be too expensive for those who’d benefit the most.

So I wondered: what if I could build something that delivered the same insights without requiring my constant presence?

Why I built this

I never set out to build a job search or career advice tool. I built it because I was tired of seeing the same broken approach to job hunting everywhere.

Most advice boils down to: “Apply more. Network harder. Fix your resume.” Then people wonder why they’re still stuck.

The whole system is designed to filter you out, not help you stand out. While you’re perfecting your resume and hitting “submit” for the hundredth time, the best opportunities are happening through a completely different channel.

I wanted to show people how the game actually works.

For years, I’ve watched the contrast between how jobs are officially filled versus how they’re actually filled. The gap is massive. People with mediocre skills but great positioning land roles while brilliant people with awful positioning stay stuck.

I’ve seen both sides. As a recruiter, I watched perfect candidates get filtered out by broken systems. As a community builder, I saw how the right connection at the right time could bypass those systems entirely. And I’ve used the same approach time and time and time again.

The solution isn’t some vague “networking” advice. It’s a specific approach to positioning yourself where opportunities naturally find you instead of you chasing them.

What I actually built

Here’s what’s wild: I built the entire system in a few days using AI.

I spent years collecting insights from hundreds of coaching conversations. Notes upon notes. Audio recordings. DMs. All my thinking about how careers actually work. I distilled this into a seven-step system that I knew would work because I’d seen it work countless times.

The technical challenge was turning this knowledge into something scalable. I didn’t want to build a static course. I wanted something that felt alive.

I used Lovable.dev to create not just the UI, but a full admin backend where I could:

  • Create and organize content for each step
  • Record audio notes directly into the system
  • Push updates instantly to users
  • Message with users directly
  • Track progress and engagement

The magic wasn’t in asking AI to generate my content. The magic was using AI to build a platform where I could deliver my content exactly how I wanted.

I first built the core design and dev spec outside of Lovable and for adding new features and ideas I’d describe what I needed: “I want users to unlock a new step every two days. Each step needs content, action items, the ability to take notes, and for me to record and attach audio notes.”

Then I’d iterate on the code until it worked exactly as I wanted.

Within days, I had a functioning product that let me upload all my curated insights and deliver them in a progressive system. Users can message me directly through the platform, creating a bridge between the scalability of software and the personal touch of coaching.

Shoutout to ÌníOlúwa Abíódún for the 4 hour late night design jam because AI definitely can’t design yet.

What surprised me

What shocked me wasn’t that AI could help build this. It was how much it changed my conception of what was possible to build alone.

A year ago, this would have required:

  • A dev team
  • A design team
  • Months of work
  • Tens of thousands of dollars

Instead, I built it myself in days. The system automatically unlocks new steps, delivers my content, plays my audio notes, and lets users message me directly if they get stuck. Fully Stripe and Supabase integration. I’ve progressively built more complex but random projects over the last few weeks, but this one felt like I learned years worth of skills in just a few days.

The first person who tried it paid $175 before I even set a price. Others are working through it and already making progress because it’s designed to push you, to take action. To do things that might feel uncomfortable.

People aren’t just paying for content. They’re paying for a structured approach that guides them through a proven system while still giving them access to me if needed.

There is still a lot to build for sure but already it’s a product that enables me to scale myself with minimal input — the core architecture is built and I just need to keep adding content and some new features as they’re needed.

What does this mean for the future of work?

The irony isn’t lost on me: I built something using AI to help people find jobs in a world where AI is changing how we work.

This shift is happening at the exact same time as massive layoffs and hiring freezes at big companies. The old path: work your way up at a prestigious company, build credentials, climb the ladder is becoming less reliable.

But that’s exactly the point. The job market has fundamentally changed, yet most job search advice remains stuck in 2010.

Meanwhile, a new path is emerging, one where visibility, thinking in public, and direct connections matter more than traditional credentials.

This is what I’ve been telling people for years, but now it’s accelerating. The good news is that there’s never been a better time to step off the application treadmill and try something different.

The bigger picture

We’re entering an era where individual creators can build sophisticated products without technical backgrounds. This mirrors what’s happening in the job market itself.

I built this system not just to help people find jobs and stand out, but to demonstrate something bigger: the tools to create value are more accessible than ever. You just have to use them.

For me, the problem was clear: the standard approach to job hunting is broken, and people need a better way.

Now I have a platform where I can share everything I’ve learned over the past decade about creating job opportunities. All my coaching philosophy, organized into progressive steps, delivered through a system I built myself.

The customers get the benefit of my thinking without requiring my constant presence. And when they do need me, they can message me directly through the system.

What excites me most isn’t just that I built this product. It’s that we’re entering a world where anyone with deep knowledge in any domain can build something similar. The barriers between having expertise and delivering it at scale are crumbling.

That’s the real opportunity I see — both for job seekers and for builders. The future belongs to people who can spot broken systems and build better ones. And now the tools to do that are in everyone’s hands.

You can check it out here — Backchannel — and if you’re building something interesting too, let me know. I’d love to learn about it.

Mindaugas

Sign up to discover human stories that deepen your understanding of the world.

Free

Distraction-free reading. No ads.

Organize your knowledge with lists and highlights.

Tell your story. Find your audience.

Membership

Read member-only stories

Support writers you read most

Earn money for your writing

Listen to audio narrations

Read offline with the Medium app

--

--

Responses (1)

Write a response