Winning First Place at the 2023 Google and Square AI Hackathon
When I first saw the announcement for the Google and Square AI Hackathon, I knew I had to participate. The challenge was clear: build something that makes e-commerce better using Square's APIs and Google's AI tools. What I didn't know was that this would turn into one of the most intense and rewarding months of development I'd ever experience.
The Idea
The inspiration came from a simple observation. Business owners are drowning in data. They have sales numbers, customer information, inventory levels, but most of them don't have the time or expertise to turn that data into actionable insights. I wanted to change that.
SquareSense was born from the idea that every business owner should be able to talk to their data. Not through complex dashboards or SQL queries, but through natural conversation. Ask your store how sales are trending. Request a chart showing your best-selling products. Have the AI explain what's happening with your customer segments.
Building SquareSense
The technical challenge was significant. I needed to integrate Square's APIs to pull real e-commerce data, then layer Google Vertex AI on top to make it intelligent and conversational. The system had to handle everything from generating interactive charts to providing psychographic analysis of customer behavior.
One of the biggest hurdles was keeping the AI's attention sharp when dealing with large datasets. A thriving e-commerce store generates massive amounts of data, and I needed the AI to maintain precision across all of it. This meant extensive fine-tuning of the Vertex AI models and rigorous stress-testing with real-world store data.
The accessibility features were particularly important to me. I researched how to make data visualization accessible to everyone, including implementing spoken AI-generated chart explanations. Data shouldn't just be available; it should be comprehensible to all users, regardless of their abilities.
The Hackathon Experience
Development moved fast. I set up authentication with Square OAuth, built the chat interface for natural language queries, implemented the chart generation system, and created the psychographic analysis engine. Each piece had to work seamlessly with the others.
The real breakthrough came when I got the system to not just show data, but explain it. Business owners could ask "Why are my sales down this week?" and get a thoughtful, data-driven answer with supporting visualizations. That felt like magic.
Judging and Results
Submitting the project was nerve-wracking. I put together a comprehensive demo video, wrote detailed documentation, and published everything to Devpost. Then came the wait.
When the announcement came that SquareSense won first place overall, I was stunned. Competing against hundreds of developers and having judges recognize the work felt incredible. The feedback highlighted how the project successfully merged AI capabilities with practical business needs.
What I Learned
This hackathon taught me that the best solutions often come from observing real problems. Business owners don't need more complex tools; they need simpler ways to access the intelligence hidden in their data.
I also learned the importance of balancing innovation with usability. Having cutting-edge AI is impressive, but if a busy store owner can't figure out how to use it in 30 seconds, it's not solving the problem.
The technical skills I developed around AI integration, real-time data processing, and accessible design have been invaluable. But more importantly, I learned to think about building products that genuinely help people, not just showcase technology.
Looking Back
SquareSense represents what I believe is the future of business intelligence: AI-driven insights that are accessible, conversational, and actionable. Winning this hackathon validated that vision and gave me the confidence to keep pushing boundaries.
If you're thinking about joining a hackathon, my advice is simple: build something you'd actually want to use. The judges can tell when you're genuinely excited about solving a problem versus just chasing prizes. Pour yourself into it, don't be afraid to tackle hard problems, and remember that every line of code is a learning opportunity.
The trophy and prize money were great, but the real win was proving to myself that I could take an ambitious idea and turn it into something real in just one month. That's a feeling I'll carry with me into every project I build.