Savora

About the project

A mobile recipe app built around one question: what can I cook with what I have? Type in your ingredients, pick a diet filter and get ranked suggestions instantly from a local catalog, enriched with results from a much larger online database when local matches run thin. The centerpiece is Linna, a Claude-powered assistant with the app's plush-sheep mascot as her face: a real conversational agent that searches your own catalog first and only reaches for the quota-limited external source when it needs to.

Key features

• Ingredient-based search with diet filters (vegan, vegetarian, with meat) and a match score that shows what you have and what is missing
• Linna, a Claude-powered assistant that calls tools: it searches the local catalog first and uses the external recipe API only when local results are weak
• Every recipe the assistant surfaces opens straight inside the app through deep navigation
• Python/FastAPI backend on Railway, rewritten from Node/Express with the original API mirrored 1:1 and covered by 73 tests before the assistant was added
• Embedded SQLite catalog that creates and seeds itself on every boot, with no manual migration step
• All external API keys stay on the server, so the client never touches a paid key
• Home screen designed around the mascot and real photography, plus favorites and a shopping list
• Built with Claude as a daily engineering partner, from the backend migration to deployment debugging