I Built My Friend an AI-Powered Allergy-Safe Meal Planner 🍳🤖

Description and use

An allergy-safe meal planner that takes user allergies and pantry items, generates three recipe suggestions via Gemini, stores history in MongoDB Atlas, and implements retry logic for AI service failures.

Problem

Meal planning becomes frustrating when you have food allergies, requiring manual checks of ingredients, pantry items, and safety.

Outcome

The project delivers a user-friendly allergy-safe meal planner that generates structured meal ideas, handles AI failures with retries, and stores history for future reference.

Workflow

• 1. User enters allergies and pantry items in the web form.
• 2. Frontend sends data to /api/spin endpoint.
• 3. Backend constructs a request to Gemini, including a system prompt and user context.
• 4. Gemini returns JSON with three meal ideas.
• 5. Backend parses and forwards the JSON to the frontend.
• 6. Frontend displays meal name, ingredients, instructions, and safety info.
• 7. Backend stores the request, response, model used, and timestamp in MongoDB Atlas.
• 8. If the AI service returns 429, 500, 503, or times out, the backend retries with exponential backoff.

Evidence

Meal planning can become frustrating when you have food allergies.

Source

The application lets a user enter: Allergies or ingredients they need to avoid; Ingredients currently available in their kitchen.

Source

The deployed version currently uses Gemini 3.8 Flash for meal generation.

Source

MongoDB Atlas is used to persist meal history.

Source

I added retry handling for temporary failures. The backend specifically handles: 429, 500, 503, Request timeouts.

Source