https://github.com/aamir-hullur/iu-hackathon-2025
A full-stack idea management platform that uses React, Vite, Firebase, and Gemini LLM to automate idea evaluation and prioritization, submitted for the IU Hackathon 2025.
https://github.com/aamir-hullur/iu-hackathon-2025
firebase gemini reactjs vite
Last synced: about 1 year ago
JSON representation
A full-stack idea management platform that uses React, Vite, Firebase, and Gemini LLM to automate idea evaluation and prioritization, submitted for the IU Hackathon 2025.
- Host: GitHub
- URL: https://github.com/aamir-hullur/iu-hackathon-2025
- Owner: Aamir-Hullur
- Created: 2025-04-15T18:58:21.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-04-15T19:59:34.000Z (about 1 year ago)
- Last Synced: 2025-04-15T20:27:47.036Z (about 1 year ago)
- Topics: firebase, gemini, reactjs, vite
- Language: JavaScript
- Homepage:
- Size: 13 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
# π Service WOW β AI-Powered Innovation Management Platform
*Service WOW* is a full-stack web platform that enables teams to submit, evaluate, and analyze innovative ideas in real time. It integrates a responsive UI with an AI-enhanced backend that automatically scores and prioritizes submitted ideas using Google's Gemini model, providing leaders with actionable insights.
> **IUB Luddy Hackathon 2025 Submission**
---
## Demo
https://github.com/user-attachments/assets/8934c27f-0573-4c18-b83c-a9bfed38c548
[Devpost Submission](https://devpost.com/software/service-wow?ref_content=user-portfolio&ref_feature=in_progress)
---
## π Project Overview
Service WOW addresses the challenge of efficiently collecting, evaluating, and prioritizing innovation ideas within organizations. Our platform uses AI to automate the scoring process, allowing teams to focus on implementation rather than lengthy manual evaluation cycles.
### π― Key Objectives
- Streamline the innovation collection process
- Provide objective, AI-driven evaluation of ideas
- Enable data-driven decision making through visual analytics
- Facilitate collaboration through voting and feedback mechanisms
- Reduce the time from idea submission to implementation decision
---
## π§ Core Features
β’β β *Idea Submission & Discovery*: Easily browse, search, and submit ideas. Toggle between grid and list views with real-time vote counts and category filters.
β’β β *AI-Driven Evaluation*: Each submitted idea is auto-scored for ROI, Effort, Strategic Alignment, and Risk using Google Gemini, including plain-language justifications and a composite score.
β’β β *Data Dashboard*: Visual dashboards show:
- ROI vs Effort with vote-based bubble sizes
- Top ideas by score
- Category-based analysis
- Priority rankings and filters
β’β β *Meeting Intelligence*: Upload β .mp3 β , β .wav β , or β .mp4 β files to extract:
- Smart summaries
- Action items with owners and deadlines
- Full transcripts
- Sentiment analysis by speaker
β’β β *Dark Mode Support*: Seamless UI theme toggling between light and dark modes.
β’β β *Interactive Features*: Real-time voting, editable metrics, and animated AI analysis visualization.
---
## π How the System Works (Interval Breakdown)
| Step | Component | Description |
|------|-----------|-------------|
| 1 | User submits a new idea via UI | Stored in Firestore β ideas_input β collection |
| 2 | Cloud Function triggers (β evaluate_idea β ) | Sends data to Gemini via structured ReAct-style prompt |
| 3 | Gemini returns enriched idea with scores and reasoning |
| 4 | Result saved to Firestore β ideas_output β collection |
| 5 | Frontend reads β ideas_output β for visualization in list and dashboard |
| 6 | User uploads meetings (optional) for analysis and summary |
β Typical end-to-end turnaround for AI scoring: *< 10 seconds*
---
## π οΈ Technology Stack
- **Frontend**: React 19, React Router 7, Tailwind CSS 4, shadcn/ui components
- **State Management**: React hooks with Firebase Firestore real-time listeners
- **Backend**: Firebase Cloud Functions with Gemini AI integration
- **Database**: Firestore with multi-collection structure
- **Styling**: TailwindCSS with dark/light mode support
- **Charts**: Recharts for data visualization
- **Forms**: React Hook Form with Zod validation
- **Deployment**: Vercel for frontend, Firebase for backend services
---
## π¦ Folder Structure (Frontend)
```bash
src/
βββ App.jsx
βββ main.jsx
βββ components/
β βββ dashboard/
β βββ ui/
β βββ FileUpload.jsx
β βββ IdeaList.jsx
β βββ NewIdeaForm.jsx
β βββ NewIdeaModal.jsx
β βββ ResultsDisplay.jsx
βββ lib/
β βββ firebaseConfig.js
β βββ utils.js
βββ pages/
β βββ AboutPage.jsx
β βββ DashboardPage.jsx
```
---
## π Getting Started
1. Clone the repository:
```bash
git clone https://github.com/yourusername/service-wow.git
cd service-wow
```
2. Install dependencies:
```bash
npm install
```
3. Set up Firebase:
- Create a Firebase project at [firebase.google.com](https://firebase.google.com)
- Enable Firestore, Cloud Functions, and Storage
- Add your Firebase configuration to `src/lib/firebaseConfig.js`
4. Deploy Cloud Functions:
```bash
cd functions
npm install
firebase deploy --only functions
```
5. Run the development server:
```bash
npm run dev
```
6. Build for production:
```bash
npm run build
```
---
## π Our Luddy Hackathon 2025 Team
- Aamir Hullur
- Atharva Gurav
- Manav Mandal
---
## π‘ Use Cases
- **Corporate Innovation Teams**: Collect and evaluate ideas from employees across departments
- **Startup Incubators**: Assess and prioritize potential startup concepts
- **Product Development**: Gather customer feedback and feature requests with AI analysis
- **Educational Institutions**: Evaluate student project proposals efficiently
- **Community Organizations**: Manage and prioritize community improvement initiatives
---
## π Project Challenges & Solutions
| Challenge | Solution |
|-----------|----------|
| AI Hallucination | Structured ReAct prompting with verification steps |
| Idea Evaluation Criteria | Multi-dimensional scoring with transparency |
| UI Performance | Optimized Firestore queries and React memo patterns |
| Media Processing | Cloud Functions with efficient chunking and processing |
| Real-time Updates | Firestore listeners with optimistic UI updates |
---
## π Acknowledgments
- IUB Luddy School of Informatics, Computing, and Engineering
- Google Firebase and Gemini API teams
- Open source community for various dependencies