Web app / live beta
I’m developing Cravate with AI assistance to help people decide what to eat from a craving, budget and the equipment available. I’ve released it as a working beta.
CravateOne craving. One answer.
One answer to “what should I eat?” matched to budget, kitchen equipment and nearby shops.
I developed Cravate with AI assistance as a single-file PWA with a Cloudflare Worker backend. I defined the product constraints and tested its recommendations against budget, time, equipment and food-safety requirements.
Open live projectCravate / development notes
In my £6 budget test, Cravate returned a £7.49 fast-food meal. I treated that as a failure: the response exceeded the budget and failed the cost and nutrition comparison.
I found that the option to recommend buying a meal had no clear conditions. The app also rejected cooking because it could not reproduce the original preparation method exactly.
My changes focus on keeping recommendations within budget, comparing cost and nutrition, and allowing simple assembly using prepared ingredients. I want the app to match the result without insisting on the same cooking method.
I still need to test whether people find the suggestions useful enough to return. Cravate is in beta, and I need real-user feedback on repeat use.
Contact