March 2026 Hackathon Recap

Introduction

We kicked off another Big Nerd Hackathon with a simple challenge: build a fully functional app in under an hour using AI.

Madi opened with a warm welcome as designers, product managers, engineers, and students from across the globe filed in. I followed with a quick overview of the rules, goals, and a few real examples to set the stage before we started building.

We had about a dozen participants building in Lovable, all competing for three awesome prizes (Lovable gift card, live user testing session, and book box set). They had about 45 minutes to brainstorm, build, and test an application while I shared tips and helped troubleshoot along the way.

In the end, we reviewed 9 standout builds that all solved real-world problems. Madi and I gave actionable feedback, awarded prizes, and, together, we all learned something and had an awesome time.

Brainstorming

Then, we moved into the core build session. First, participants registered for Lovable if they didn’t already have an account. Then, they took a few minutes to come up with an idea for an app. It could be anything they wanted. Bonus points if:

• It solves a real-world problem
• It helps humans in a meaningful way
• It’s an original idea
• You’ve followed the directions

They kept the scope intentionally small.

Prompting

Over the next few minutes, they completed the following template (using their brainstorming ideas) as their Lovable build script:

Building

They then copied and pasted their completed build scripts directly into the Lovable prompt and sat back while Lovable built their apps.

Caption: This screenshot shows copy/pasting a build script into the Lovable prompt to build an app that helps first-time home buyers.

Testing

When Lovable finished building (about 5 minutes), they tested their apps manually using the following guidelines:

• Don’t worry about visual polish yet
• Focus on core functionality
• A couple of small bugs are okay
• A broken workflow is not
  

They tested manually to avoid credit burn and focused on fixing anything that broke core functionality.

Caption: This screenshot shows live testing of an app that helps first-time home buyers understand how to care for their home.

Sharing

When participants were done testing and felt comfortable with their builds, they raised their hands (in Google Meet) to indicate they were ready to demo.

They had about 3-4 minutes each to share what they built and had to answer the following questions:

• Who is it for
• What problem does it solve?
• What is the core functionality? (show us)

Caption: This screenshot shows a participant (Jaden) who created an app called CalmKids for children and parents.

Judging

Our AI-powered Idea Inspector agent judged the hackathon in real time. The agent recorded the demos and objectively evaluated them against the criteria in the instructions and a 8-factor rubric that I created, which includes the following weighted categories:

• Real-World Problem (20%)
• Human Impact (20%)
• Originality (15%)
• Usability (15%)
• Feasibility (10%)
• Scalability (5%)
• Ethics (15%)

The judge also provided feedback in the form of strengths and areas for improvement.

Finally, the app keeps track of the scores through a leaderboard, making it easy for Madi and me to track results. In the event of a tie, we can also apply human judgment to determine the outcome.

Reflection

The hackathons continue to be a huge success. We aim to create an engaging and welcoming environment and provide a ton of value, a safe space to learn, share, and ask questions, and the opportunity to walk away with an app you’re proud of, and potentially an awesome prize.

“It was a great learning experience. Hoping for more such sessions!”

Mitaali

We hope you join us at the next free event and look forward to seeing what you build next!