Eat with GDM

Why I built this

When I was pregnant with my daughter, I was diagnosed with gestational diabetes. I had to log all my food and monitor my glucose levels after every meal, and doing it in a spreadsheet was a real pain. This time around, I know my odds of getting gestational diabetes again are high, so I used AI to build myself a simple app to make managing it just a little more pleasant.

Eat with GDM has a timer to remind me to check my glucose an hour after every meal, lets me log food by typing or voice, and looks up carb counts against a USDA database so I’m not doing mental math while I eat. I even built a coach into it to help me brainstorm GDM-friendly food pairings. I’m still iterating (I use it about six times a day!), but it feels good enough to put out as a beta.

The usual disclaimer applies: I’m not a medical expert, and this isn’t a substitute for advice from your own doctor. It’s just the app that’s helping me manage my own diagnosis.

What it does

• Log fasting and post-meal glucose readings, tagged as test strip or CGM.
• A built-in timer reminds you to check your glucose an hour after each meal.
• Log meals and snacks by typing or voice, with a walk-after-eating toggle.
• Look up carbs for common foods against a USDA food database directly in the app instead of having to go to another search engine.
• An AI coach to help brainstorm GDM-friendly food pairings.
• Export your log as a clean, doctor-shareable CSV or PDF for appointments.

Screenshots

Try it

Eat with GDM is currently in beta on TestFlight for iOS. I’m still iterating daily (I use it myself, multiple times a day), so expect rough edges while I keep polishing it.

Not medical advice - always follow your own doctor’s guidance for managing gestational diabetes.

How I built it

This is a native iOS app, built with SwiftUI and local-first storage, since keeping health data private and safe from loss mattered more here than anything else. No accounts, no sync, no analytics on health data - everything stays on-device unless you explicitly export it. The one exception is the coach: it sends some context to a small proxy service I built, so it can give AI-assisted food suggestions. The carb lookup is backed by the USDA FoodData Central database.