DEV Community

Cover image for Green Habit Coach — AI-Powered Habit Analysis and Eco Coaching
Zoe Lin
Zoe Lin

Posted on

Green Habit Coach — AI-Powered Habit Analysis and Eco Coaching

DEV Weekend Challenge: Earth Day

What I Built

I built Green Habit Coach, a web app that helps users reflect on their daily habits and get practical eco-friendly suggestions.

Users can log in, fill out a short habit form, and receive an analysis result based on their lifestyle inputs. The app looks at habits such as transportation, meat consumption, air conditioning usage, disposable item usage, recycling behavior, and bring-your-own habits.

After the analysis, users get an eco score, a short summary, top issues, suggestions, and a simple 7-day challenge plan. They can also continue the experience by asking follow-up questions through an AI coach, such as what to improve first this week or which habit is the easiest to start tomorrow.

I wanted to make sustainability feel more personal, more practical, and easier to act on in everyday life.

Demo

Code

How I Built It

I built the frontend with React + Vite and the backend with FastAPI.

On the frontend, users can log in, submit their habit data, view their latest analysis, load history records, and ask follow-up questions in the coach section.

On the backend, the app handles authenticated API requests, habit analysis, history storage, coach thread creation, and coach follow-up messaging.

The app uses:

  • Auth0 for login and protected API access
  • Backboard for the follow-up AI coach experience
  • SQLite + SQLAlchemy for storing analysis history
  • Firebase Hosting for the frontend
  • Google Cloud Run for the backend

One of the biggest parts of this project was not just building the features, but also making the whole flow work in deployment. I spent a lot of time debugging Auth0 configuration, CORS between the frontend and backend, Cloud Run deployment issues, and async backend integration for the coach flow.

That was actually one of the most valuable parts of the project, because it made the app feel like a real full-stack product instead of only a local demo.

Prize Categories

I’m submitting this project for:

  • Best Use of Backboard
  • Best Use of Auth0 for Agents

Backboard powers the coach conversation flow and makes it possible to continue the user journey after the initial habit analysis.

Auth0 helps connect the experience to real authenticated users, protect backend routes, and keep history data scoped to the correct user.

What I Learned

This project helped me practice real full-stack development, especially:

  • frontend/backend integration
  • cloud deployment
  • authentication
  • production debugging
  • building AI features that feel useful

It also gave me more experience turning a simple idea into a deployed product with a complete user flow.

Top comments (0)