This is a submission for the World's Largest Hackathon Writing Challenge: After the Hack.
So I built this thing called Metric Moon for the World's Largest Hackathon. Basically, you ask your data questions in plain English and get instant, visual answers. Pretty standard AI stuff, right? I thought I was done with it after submission.
Then last week I was grabbing coffee with Maria (she runs this small animal rescue), and she's pulling her hair out over a laptop full of spreadsheets. Trying to figure out which adoption campaigns work best, what times of year they get more surrenders, basic stuff that could help them save more animals.
"I know the patterns are in here," she says, scrolling through endless rows. "But learning Excel formulas is like learning a foreign language, and we're all volunteers."
And I'm sitting there thinking... wait. This is exactly what Metric Moon does. She could just ask "which adoption events had the highest success rate?" and get an actual answer instead of spending her Saturday afternoon fighting with VLOOKUP functions.
The thing that's been bugging me
It's not that Maria isn't smart - she's brilliant at what she does and has saved hundreds of animals. But she's locked out of her own data because the tools assume you either have a CS degree or a $50k budget.
I used to think this was just "how things work." You want insights? Learn a query language/s. Can't learn it? Hire someone. Can't afford that? Too bad.
But now I'm wondering why we accept that. Like, we figured out how to make smartphones intuitive enough that anyone's grandparents can video chat with their grandkids. Why are we still gatekeeping data behind technical barriers?
What I actually learned building this
The technical stuff was honestly pretty straightforward. SkyAI Agents does most of the heavy lifting, React and Recharts handle the visuals, nothing revolutionary. I started with space mission data because it seemed like a cool demo dataset.
But what caught me off guard was realizing how many people like Maria are out there. People who have valuable data but can't do anything meaningful with it because the learning curve for data tools is just brutal.
It's not that the technology is missing; we have incredible databases, powerful analytics tools, and now sophisticated AI models. The problem is that none of it is accessible unless you already speak the technical language.
Where I'm going with this (maybe)
I keep going back to Metric Moon, not because I have some grand business plan, but because I can't stop thinking about Maria and her spreadsheets.
The space missions dataset was fine for a demo, but what if this worked with any database? What if Maria could upload her adoption data and just... ask it questions? What if my food truck friend could connect his POS system and improve his sales?
I don't know if this turns into anything bigger. Maybe I'm overthinking a weekend project. But I keep coming back to this idea that access to your own insights shouldn't require a technical degree.
The messy reality
Look, I'm not trying to put data analysts out of work or pretend AI solves everything. Complex analysis still needs experts. But asking basic questions about your own data? That shouldn't be rocket science.
The hackathon got me thinking about who actually gets to benefit from the stuff we build. Most hackathon projects die after submission, but this one keeps poking at me.
Maybe that means something. Maybe it doesn't. But I'm going to keep working on it and see what happens.
Want to see Metric Moon in action? Check out the demo and my original technical deep-dive.
Top comments (0)