FeedPulse — AI-Powered Product Feedback Platform

FeedPulse is a lightweight internal tool that helps teams collect product feedback and feature requests, then uses Google Gemini AI to automatically categorize, prioritize, and summarize them—so product teams can quickly understand what to build next.

---

Overview

Product feedback often lives across emails, chats, support tickets, and spreadsheets. FeedPulse centralizes feedback into a single system and turns raw input into actionable insights with AI-powered classification and summarization.

Use cases

• Collect feature requests from users/customers
• Triage and prioritize feedback automatically
• Generate summaries for product planning and roadmap discussions

---

Features

• Feedback collection: Capture product feedback and feature requests in one place
• AI categorization: Automatically groups feedback into meaningful categories
• AI prioritization: Helps identify high-impact requests (based on volume/importance signals)
• AI summarization: Generates concise summaries for quick review
• Secure API: Authentication with JWT, password hashing with bcrypt
• Operational safety: Rate limiting, security headers, request logging

---

Tech Stack

Frontend

• Next.js (React + TypeScript)
• Tailwind CSS
• Axios

Backend

• Node.js + Express (TypeScript)
• MongoDB + Mongoose
• Authentication: JWT + bcrypt
• Security: Helmet, Express Rate Limit
• AI: Google Gemini (@google/generative-ai)
• Logging: Morgan
• Config: dotenv

---

System Architecture

┌───────────────────────────────────────────────────────────┐
│                     Frontend (Next.js)                    │
│  UI for submitting feedback + viewing AI insights         │
└───────────────────────────────────────────────────────────┘
                          │  HTTP (REST)
                          ▼
┌───────────────────────────────────────────────────────────┐
│                      Backend (Express)                    │
│  Auth (JWT) • Feedback APIs • AI processing (Gemini)      │
└───────────────────────────────────────────────────────────┘
                │                         │
                │ Mongoose                │ Google Gemini API
                ▼                         ▼
┌─────────────────────────────┐     ┌───────────────────────┐
│      MongoDB (Database)     │     │  Gemini AI (External)  │
│ feedback, users, metadata   │     │ classify/summarize     │
└─────────────────────────────┘     └───────────────────────┘

---

Project Folder Structure

This is the expected high-level structure for this monorepo (frontend + backend).
Folder/file names may vary depending on your implementation.

FeedPulse/
├── backend/
│   ├── src/
│   │   ├── server.ts              # App entry point
│   │   ├── routes/                # API routes
│   │   ├── controllers/           # Request handlers
│   │   ├── services/              # Business logic (Gemini processing, etc.)
│   │   ├── models/                # Mongoose models (User, Feedback, etc.)
│   │   ├── middleware/            # Auth, validation, rate limit, error handling
│   ├── package.json
│   ├── tsconfig.json
│   └── .env                       # Not committed (local)
│
├── frontend/
│   ├── app/                       # Next.js App Router pages
│   ├── components/                # UI components
│   ├── lib/                       # API client, helpers
│   ├── public/                    # Static assets
│   ├── package.json
│   └── .env.local                 # Not committed (local)
│
├── README.md
└── .gitignore

---

Installation

Prerequisites

• Node.js (LTS recommended)
• npm (or pnpm/yarn)
• MongoDB (local or hosted)
• Google Gemini API key

Clone the repository

git clone https://github.com/supunakalanka76/FeedPulse.git
cd FeedPulse

Install dependencies

Backend

cd backend
npm install

Frontend

cd ../frontend
npm install

---

Configuration

This repo may not include .env.example yet. Create the following env files locally.

Backend environment (backend/.env)

Create backend/.env:

PORT=4000
MONGO_URI=mongodb+srv://<db_username>:<password>@cluster0.yhoye6q.mongodb.net/feedpluse?retryWrites=true&w=majority

JWT_SECRET=replace_with_a_strong_secret

# Google Gemini
GEMINI_API_KEY=your_gemini_api_key_here

# Optional (useful for CORS)
FRONTEND_URL=http://localhost:3000

Frontend environment (frontend/.env.local)

Create frontend/.env.local:

NEXT_PUBLIC_API_BASE_URL=http://localhost:4000

If your backend serves routes under a prefix like /api, use: NEXT_PUBLIC_API_BASE_URL=http://localhost:4000/api

---

Run the Application

Development mode

Start backend

cd backend
npm run dev

Backend should run on: http://localhost:4000

Start frontend (new terminal)

cd frontend
npm run dev

Frontend should run on: http://localhost:3000

Production mode (optional)

Backend

cd backend
npm run build
npm start

Frontend

cd frontend
npm run build
npm start

---

Screenshots

Add screenshots to a folder like docs/screenshots/ and link them here.

Example:

• docs/screenshots/admin_dashboard.png
• docs/screenshots/feedback_form.png

Public Feedback Page

Admin Dashboard

Login Page

---

Testing

If you haven’t added tests yet, this section can remain as a placeholder.

Backend

cd backend
npm test

Frontend

cd frontend
npm test

---

Contribution

Contributions are welcome.

1. Fork the repo
2. Create a branch: git checkout -b feature/my-feature
3. Commit: git commit -m "Add my feature"
4. Push: git push origin feature/my-feature
5. Open a Pull Request

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Made with ❤️ by Supun Akalanka

---