Mood Meal: Mood-Based Meal Recommendation App
A live, team-built web app that suggests meals based on mood — originated by friends as a personal project, with the majority of the technical build led by me
01 Problem & Context
Deciding what to eat can take more thought than it should, especially once mood is factored in. Mood Meal — an idea originated by friends as a personal project — set out to solve this with a simple, mood-driven meal recommendation app rather than another generic recipe search tool.
02 Objectives & Constraints
- — Build a full-stack app connecting user mood to personalized meal suggestions
- — Implement authentication (signup/login) with persistent user data
- — Ship a working, deployed product rather than a prototype
03 Process
Process
From Idea to Architecture
The idea for Mood Meal came from friends as a personal project — a simple question of “what should I eat, given how I feel” turned into a scoped product idea. Translating that into a working system meant making most of the concrete technical decisions independently: a Next.js frontend, a FastAPI backend, and Supabase for authentication and data storage, wired together and deployed to Vercel (frontend) and Render (backend).
Build and Deployment
AI-assisted development tools helped accelerate parts of the implementation, but the architecture, data model, and day-to-day technical problem-solving were my responsibility. The result is a live, deployed application rather than a local prototype — handling account creation, login, and mood-based meal suggestions end to end.
04 Challenges & Solutions
Translating a friends' product idea into a working technical architecture required making most of the concrete technical decisions independently.
Selected and integrated a Next.js + FastAPI + Supabase stack, deployed on Vercel and Render, and used AI-assisted development tools to accelerate implementation while retaining full ownership of the architecture and problem-solving.
05 Results & Outputs
- → Live, deployed application (mood-meal-project.vercel.app) handling user authentication and mood-based meal recommendations
06 Lessons Learned
- → Leading the technical execution of someone else's idea clarified the difference between owning a concept and owning an implementation — most of the architecture, backend, and database decisions were mine even though the idea wasn't.
- → AI-assisted tools sped up implementation, but the judgment calls — architecture, data model, what to build first — still had to be mine.