A hackathon-winning AI-powered career advisor platform specifically designed for Indian students. Built with React, Firebase, and Google Gemini AI.
Disha AI provides hyper-personalized career guidance using Google's cutting-edge AI technology. The platform analyzes students' interests, academic strengths, and existing skills to recommend career paths tailored for the Indian job market.
- Seamless Onboarding: Anonymous sign-in with intuitive multi-step wizard
- AI-Powered Recommendations: Google Gemini API generates personalized career suggestions
- Skill Gap Analysis: Visual comparison between required and existing skills (green for owned, blue for needed)
- Detailed Roadmaps: Comprehensive learning paths with Indian resources (NPTEL, Swayam, etc.)
- Mobile Responsive: Fully optimized for all device sizes
- Real-time Updates: Dynamic content generation and caching
- Frontend: Vite + React 18 + JavaScript
- Styling: Tailwind CSS with custom design system
- Backend: Firebase (Firestore + Cloud Functions)
- Authentication: Firebase Auth (Anonymous)
- AI Integration: Google Gemini API (gemini-2.0-flash-exp)
- Animations: Framer Motion
- Hosting: Firebase Hosting
- Node.js (v18 or higher)
- Firebase CLI
- Google Cloud Platform account
- Firebase project
git clone <repository-url>
cd disha-ai
npm install# Install Firebase CLI
npm install -g firebase-tools
# Login to Firebase
firebase login
# Initialize Firebase project
firebase init
# Select:
# - Hosting
# - Functions
# - FirestoreCreate .env file in the root directory:
cp .env.example .envAdd your Firebase configuration:
VITE_FIREBASE_API_KEY=your_api_key_here
VITE_FIREBASE_AUTH_DOMAIN=your_project_id.firebaseapp.com
VITE_FIREBASE_PROJECT_ID=your_project_id
VITE_FIREBASE_STORAGE_BUCKET=your_project_id.appspot.com
VITE_FIREBASE_MESSAGING_SENDER_ID=your_sender_id
VITE_FIREBASE_APP_ID=your_app_id# Configure Gemini API key for Cloud Functions
firebase functions:config:set gemini.api_key="your_gemini_api_key"cd functions
npm install
cd ..
firebase deploy --only functionsnpm run build
firebase deploy --only hosting# Start development server
npm run dev
# Start Firebase emulators (optional)
firebase emulators:startnpm run dev- Start Vite development servernpm run build- Build for productionnpm run preview- Preview production buildnpm run firebase:serve- Serve with Firebase hostingnpm run firebase:deploy- Deploy to Firebase
disha-ai/
βββ src/
β βββ components/
β β βββ LandingPage.jsx # Hero section & landing
β β βββ OnboardingWizard.jsx # Multi-step profile creation
β β βββ Dashboard.jsx # Main user dashboard
β β βββ CareerCard.jsx # Individual career recommendations
β β βββ RoadmapModal.jsx # Detailed roadmap with skill gap analysis
β βββ hooks/
β β βββ useAuth.js # Firebase authentication hook
β βββ services/
β β βββ firebase.js # Firebase configuration
β β βββ userService.js # Firestore operations
β βββ App.jsx # Main app component
β βββ main.jsx # React entry point
β βββ index.css # Tailwind + custom styles
βββ functions/
β βββ src/
β βββ index.ts # Cloud Functions with AI integration
βββ public/
βββ firebase.json # Firebase configuration
βββ tailwind.config.js # Tailwind customization
βββ vite.config.js # Vite configuration
- Primary: Deep Indigo (#6366f1) - Professional, trustworthy
- Accent: Emerald Green (#34d399) - Skills you have
- Electric: Electric Blue (#60a5fa) - Skills to learn
- Background: Light gray (#f9fafb) - Clean, minimal
- Font: Inter (clean, modern, readable)
- Headings: Bold, gradient text effects for impact
- Body: Medium contrast, optimized for readability
The platform's most innovative feature visually compares required skills vs. user skills:
- β Green tags: Skills the user already has
- π― Blue tags: Skills the user needs to learn
- π Progress bar: Overall skill match percentage
- π― Interactive: Click to explore detailed roadmaps
- Secure: Gemini API calls only through Cloud Functions
- Structured: JSON-based prompts for consistent outputs
- India-focused: Prompts specifically mention Indian market context
- Caching: Generated content saved to Firestore for performance
- Anonymous authentication for frictionless onboarding
- Firestore security rules prevent unauthorized access
- API keys properly secured in Cloud Functions
- Client-side validation with server-side verification
- Fully responsive design using Tailwind CSS
- Touch-optimized interactions
- Optimized loading states and animations
- Progressive enhancement approach
- Vite: Lightning-fast development and builds
- Code splitting: Lazy loading of components
- Optimized images: Properly sized and compressed
- Caching: Firebase CDN + intelligent data caching
- Modern tech stack (React 18, Firebase, Gemini AI)
- Clean, well-documented code
- Proper error handling and loading states
- Responsive design and performance optimization
- Visual skill gap analysis (unique feature)
- AI-powered personalization
- Seamless anonymous onboarding
- Real-time content generation
- Specifically targets Indian students
- Includes Indian resources (NPTEL, Swayam)
- Addresses education-to-career pipeline
- Scalable solution for mass impact
- Clear monetization paths (premium features, partnerships)
- Scalable serverless architecture
- Low operational costs
- Large addressable market (Indian students)
- Intuitive, wizard-based onboarding
- Beautiful, clean interface
- Smooth animations and transitions
- Mobile-first design
This project is built for the Google Gen AI Exchange Hackathon.
For contribution raise the PR For questions or collaboration opportunities, please reach out to me.
Disha AI - From Career Confusion to Crystal Clarity πβ¨