His Eyes Were Red After Every Recess. So I Built a Pollen Map.
The messages from my son's teacher kept coming, always after recess. His eyes were swollen and red. His nose wouldn't stop running. He couldn't focus. She was worried. I was worried. We thought it was a cold, then we thought it was allergies, then we weren't sure what it was anymore.
By the time we got him to the doctor, his sinuses were so congested that the mucus had backed up into his ear and throat. What looked like "just bad allergies" was an ear infection. Antibiotics. A daily allergy medicine. And a new problem I didn't know I had: now I had to decide, every single morning, how much medicine a small kid actually needs.
I needed the answer, not a weather app
The medical issue wasn't the hard part. We had a plan. The hard part was judgment. Some days the pollen is bad enough that he needs the full dose. Some days it's mild and layering allergy meds on top of a small kid feels like overkill. I didn't want him on more medication than he needed. I also didn't want him back in the nurse's office on Monday afternoon.
What I wanted felt simple. A clear picture of what's actually in the air where we live, plus a daily read on how heavy it is. The weather apps give you one number. The allergy sites give you a region. Neither gives you the thing I actually needed, which was more specific. Which plants. Which days. Which ones are climbing.
The version that already exists isn't built for this
I started looking around. There's plenty of pollen data out there, between the National Allergy Bureau, the state forecast, the university extension services, the weather apps, and the general allergy sites. Each one shows a slice. None of them put it together in a way that made sense for a mom with a kid who has a known reaction and a finite tolerance for side effects.
I spent an evening trying to triangulate between three sources and realized I was doing exactly what I help corporate teams stop doing: hunting for answers across fragmented tools because no one built the thing that matched my actual question. Then I realized I didn't need to wait for someone else to build it.
What I built in a weekend
So I stopped hunting and started building. With Claude's help, I pulled the data I cared about into one map. Charlotte-specific. The top allergens for our area. A daily read on severity. A view of what's going up and what's coming down. Then I took it into Canva to clean up the visuals so I could actually use it on a phone at 7 a.m. while also packing a lunch.
It's not fancy. It's not polished. It's a tool built for one family's weird morning math, and that's exactly what makes it useful. Now, when I'm deciding whether my son needs the full dose today, I'm not guessing.
I'm not a biologist. I don't need to be.
Here's the part I keep coming back to. I couldn't have built this a few years ago. I'm not an allergist. I'm not a meteorologist. I don't have time to learn a new scientific discipline to solve a small family logistics problem.
I didn't have to. I already knew the question I needed answered. The specialized knowledge (the pollen taxonomy, the forecasting models, the visualization choices) could be pulled together by an AI that knows how to reach across those domains. My job was to know what I actually needed, in my life, for my kid. The tool helped me build it.
This is where a lot of conversations about AI get stuck. We talk about it as if the goal is becoming more expert. Learn to prompt better. Learn the models. Learn the APIs. Those things matter if you build products for a living. For everyone else, most of what AI can do has nothing to do with mastery. It has to do with assembly.
What AI is actually good at, for the rest of us
The future isn't every parent becoming an amateur data scientist. The future is every parent, every teacher, every caregiver, every person with a small problem no one else noticed, being able to pull the right pieces together without having to understand all of them.
I needed a specific map for a specific kid in a specific zip code. I got one. It took a weekend. It costs me almost nothing to run. And it takes a real worry out of my morning.
That's AI adoption, for me. Not a transformation. Not a revolution. A small tool that matches the shape of an actual problem, built by someone who knew the problem cold but didn't need to know everything else.
What's changed since I built it
My son still takes his allergy meds. I still have a runny nose most mornings. The map doesn't fix allergies. But it does something I didn't have before. It gives me a way to match the dose to the day instead of defaulting to "more" because I'm anxious.
That tradeoff, between anxious over-action and an informed decision, is where a lot of family life actually happens. It's where a lot of work life happens too. We act bigger than we need to because we don't have the signal we need. A tool that gives you the signal quietly changes how you show up.
That's what I want to keep building. Small tools for specific problems, assembled from knowledge that used to be locked behind specialists. Not to replace the specialists. To give ordinary life a little more precision.
If you want to see what's blooming in Charlotte this week, the map is above. If you live somewhere else and have a kid going through the same thing, I'd say build one. You don't need to know biology. You just need to know your kid.
Comments