We didn't win. But somewhere between hour 14 and hour 21, I think we found something better.
The 24-Hour Clock Starts
There's a strange electricity in a hackathon room at midnight. Fifty-something students hunched over laptops, energy drinks stacking up like tiny monuments to ambition. The theme had just been announced: "Mending Over Ending."
My team (four of us) immediately started brainstorming. We had 24 hours to build something real, something that spoke to the idea of fixing what's broken instead of throwing it away.
We slept 3 of those hours. We coded for the other 21.
This is the story of how we built Talent Bridge, a web app that reframes layoffs not as career endings, but as transitions.
The Original Idea (And Why We Killed It)
Our first idea was cinematic. We wanted to build a decision-based game set 60 seconds before the end of the world. Every choice the player made would branch the story into a different ending. It was dramatic, creative, and honestly, kind of fun to think about.
But as we sketched it out, something didn't click. We kept coming back to the theme. "Mending Over Ending" wasn't about dramatic apocalypse endings. It was about what happens after something breaks. It was about recovery. Rebuilding. Second chances.
So around hour 3, we pivoted.
We scrapped the game entirely and asked ourselves a different question: What's actually ending for real people right now?
The answer was obvious. In 2024 and 2025, the tech industry saw massive layoffs: hundreds of thousands of engineers, designers, and PMs let go from companies they'd given years to. For many of them, it felt like an ending. But it didn't have to be.
The Pivot: From Game to Platform
We built Talent Bridge: a web application that connects laid-off workers from large companies with hiring opportunities at smaller companies that need their skills.
The idea was simple but meaningful:
- Big companies lay off talented people. Those people have experience, domain knowledge, and skills.
- Small companies struggle to hire. They can't compete with FAANG salaries, but they can offer meaningful work, ownership, and growth.
- Talent Bridge connects them. It "mends" careers that might otherwise "end."
The theme wasn't a metaphor for us. It was the product itself.
Building at 3 AM with Cursor
Here's where it gets technical (and honestly, this was the most eye-opening part of the hackathon for me).
We used Cursor, an AI-assisted coding editor, as our primary development tool. And it fundamentally changed how we worked under pressure.
In a hackathon, speed is everything. You don't have time to carefully architect your codebase or write perfect tests. You need to move fast, break things, and fix them faster. Cursor fit perfectly into that flow.
Here's what the process actually looked like:
- Scaffold fast. We described components and Cursor generated the boilerplate. We'd tweak, iterate, move on.
- Debug in real time. When something broke at 2 AM and our brains were running on fumes, being able to highlight a block of code and ask "why isn't this working?" saved us more than once.
- Vibe, fix, vibe, fix. That became our rhythm. We'd get into a creative flow, hit a wall, use AI to unstick ourselves, and keep moving. It wasn't about replacing our thinking; it was about compressing the feedback loop.
The experience taught me something important: AI-assisted development isn't about writing less code. It's about spending more time on the decisions that matter: the architecture, the user experience, the story you're telling with your product.
We still wrote plenty of code by hand. We still argued about design choices at 4 AM. But Cursor let us stay in flow when the clock was ticking.
The Demo
Presenting after 21 hours of building and 3 hours of sleep is an experience I won't forget. Your eyes are burning. Your voice is hoarse. But there's this adrenaline that kicks in when you see your project on the screen and realize: we actually built this.
We demoed Talent Bridge live. The matching system worked. The UI was clean enough to be presentable. We walked through the concept, the pivot story, and showed how the app connected displaced workers with companies looking to hire.
The judges asked good questions. We answered honestly: about what worked, what we'd improve, and why we pivoted from the original idea.
We Didn't Win
I'm not going to pretend that didn't sting a little. When you pour 21 hours into something and barely sleep, you want validation. That's human.
But here's what I realized walking out of that room: the scoreboard wasn't the point.
We came in with a vague game idea. We left with a working web application that addressed a real problem. We practiced building under extreme time pressure, collaborating when everyone was exhausted, and making a hard pivot when the first idea wasn't strong enough.
That takes more courage than sticking with a safe plan.
What I Actually Learned
Technically:
- How to build a functional web app end-to-end in under a day
- How AI-assisted development tools like Cursor can accelerate iteration without replacing critical thinking
- The importance of scoping ruthlessly: cutting features that don't serve the core idea
- How to structure a codebase for speed without creating unmaintainable chaos
Personally:
- Pivoting isn't failure. It's clarity.
- Sleep deprivation is a terrible long-term strategy but a surprisingly effective bonding experience
- The best ideas come from asking "what actually matters?" instead of "what sounds cool?"
- Losing doesn't erase the experience. The skills, the memories, and the project itself: those are yours to keep.
Closing Thought
The theme was "Mending Over Ending." We built a product about mending careers. But in a way, the hackathon itself was about mending my own assumptions: about what winning means, about how I work under pressure, and about what I'm capable of building when the clock is against me.
Talent Bridge lives on GitHub. It's not perfect. It's a 21-hour hackathon project. But it's real, it works, and it stands for something I believe in: that endings are just transitions waiting for the right bridge.
If you're thinking about joining a hackathon: do it. You might not win. But you'll walk away with something no trophy can give you: proof that you can build under pressure, adapt when plans fall apart, and still ship something you're proud of.
- Minh