Inspiration
It started with two small disasters. Eisha was traveling in Europe and forgot her medicine. She is allergic to ibuprofen, so she couldn't just buy whatever was on the pharmacy shelf. Beline forgot her charging cable on a trip and had to buy an emergency one at a steep price.
Neither of us is disorganized. We just carry too much in our heads. Research backs this up. Sociologist Allison Daminger calls it cognitive labor: anticipating needs and monitoring what is running out. In her study, the woman carried more of it in 26 of 32 couples.
She doesn't only carry the bag. She carries the list.
What it does
Chelsea is a bag manager for women. It knows what is in every bag she carries: makeup, diaper, travel and everyday. It predicts when each item will run out or expire. It sends a calm reminder before that happens, with a reorder link.
If an item is out of stock, Chelsea suggests three alternatives, each with one line on why it is similar. Search respects her allergies, diet and values by default. A trip planner builds her packing list and flags what to buy before leaving. Every Friday, a Surprise Bag offers four small things chosen for her: a product that pairs with one she owns, a routine using what she has, a small treat and a useful tip.
How we built it We started from the problem, not the product. We used two personas: Sofia, a working mom, and Maya, a mom flying with two toddlers. We wrote a full build spec first. It covers the screens, the data model, the logic and a strict design system. The app is built with Expo, React Native and TypeScript. It runs as a web app on Vercel, designed for iPhone. The logic is rule-based for now: Run-out forecasts use size, usage and opening date. Search filters out avoid-list ingredients first, then ranks by fit. The Surprise Bag is fixed for each week. The design is editorial: serif type, white space, and products floating over an open tote. No gradients, no alarms, no "AI magic" wording. Challenges we ran into Forecasting without data. We don't know her real usage on day one. So we added one-tap corrections: "I'm almost out" and "Still plenty." Calm, not more noise. An app about mental load must not add to it. We cut red alerts, streaks and guilt. Smart without looking "AI." Recommendations had to read like editorial curation. Scope. We had to fit a full, clickable demo into hackathon time, on mock data. Brands. We used fictional products and original illustrations, with no real logos. Accomplishments that we're proud of Every tab works on an iPhone, with no dead buttons. A search for "vegan snack" returns results that explain why they fit, for example "Vegan and nut-free, like the bars in your diaper bag." The Surprise Bag opening animation. A product grounded in research on mental load, not just in shopping. What we learned
The hardest part of carrying a bag is the noticing, not the carrying. Existing tools all start after she has already noticed. We also learned that wellbeing is holistic. It covers mind, body, joy, values and money at once. Finally, design is part of care: a calm interface is itself a wellbeing feature.
What's next for Chelsea - your bag manager Backend. Supabase for accounts and saved bags, with data locked per user. Real reminders. A daily job runs the forecasts and sends push and email alerts. Photo scanning. A vision model identifies what is in a photo of her bag. Live catalog. Real products, availability checks and reorder links. Pilot study. We will measure fewer run-outs, minutes saved and self-reported mental load.
Built With
- claude
- html
Log in or sign up for Devpost to join the conversation.