🚀 RunaGen AI Career Companion

Google Cloud Gen AI Exchange Hackathon 2025

A comprehensive career development platform powered by Google Cloud's Gemini AI, featuring intelligent resume analysis, personalized learning roadmaps, and interactive career simulations.

Live Demo

RunaGen AI Prototype — Live Link

RunaGen AI Prototype — Demo Video Link

RunaGen AI Prototype — Presentation Deck

Note: The backend currently runs on http://localhost:3001. To fully test the prototype, open the project in VS Code and run the backend locally as described in the Quick Start section.

🎯 Project Overview

AI Career Companion is an innovative career development platform that leverages Google Cloud's Gemini AI to provide personalized career guidance, skill gap analysis, and interactive learning experiences. The platform helps users identify career opportunities, create personalized learning roadmaps, and practice real-world scenarios through AI-powered simulations.

🏆 Hackathon Focus

This project demonstrates the power of Google Cloud's Generative AI services, specifically:

Gemini 2.5 Flash for intelligent content generation
Vertex AI for advanced AI capabilities
RAG (Retrieval-Augmented Generation) for context-aware responses
Vector Search for semantic understanding

🛠️ Tech Stack

Frontend

React 18 with TypeScript
Vite for fast development and building
Tailwind CSS for modern, responsive UI
Lucide React for beautiful icons
Three.js for 3D visualizations

Backend

Node.js with Express.js
ES Modules for modern JavaScript
MongoDB with Mongoose ODM
Multer for file uploads
PDF-Parse for document processing

AI & Cloud Services

Google Cloud Vertex AI (Gemini 2.5 Flash)
YouTube Data API v3 for video recommendations
Text Embeddings for semantic search
Vector Store for RAG implementation

Development Tools

ESLint for code quality
PostCSS with Autoprefixer
TypeScript for type safety

✨ Key Features

🎯 Working Features

1. AI-Powered Resume Analysis

✅ PDF/Word/PowerPoint/Image parsing with intelligent text extraction
✅ Gemini-powered skill extraction using RAG
✅ Job role auto-detection with confidence scoring
✅ Skills gap analysis with priority-based recommendations
✅ Job matching with compatibility scores

2. Personalized Learning Roadmaps

✅ 3-stage skill gap analysis (Critical, Important, Nice-to-have)
✅ YouTube video integration with real-time search
✅ Exam preparation with certification recommendations
✅ Practice projects with skill development tracking
✅ Learning platform suggestions (Coursera, Udemy, etc.)

3. AI Career Mentor

✅ Conversational AI mentor powered by Gemini
✅ Context-aware responses using RAG
✅ Resume analysis integration for personalized advice
✅ Conversation history with MongoDB persistence
✅ Badge system for engagement tracking

4. Career Simulations

✅ Interactive scenarios for different career paths
✅ Skill-based challenges with real-time feedback
✅ Progress tracking and completion metrics
✅ Simulation templates for various roles

5. Data Persistence

✅ MongoDB integration for all user data
✅ File storage with automatic cleanup
✅ User session management
✅ Conversation history with search capabilities

⚠️ Features with Limitations

1. YouTube Integration

⚠️ Mock data fallback when API key not configured
⚠️ Rate limiting on YouTube API calls
✅ Real video links when API key is provided

2. AI Services

⚠️ Fallback mechanisms when Gemini API is unavailable
⚠️ Error handling with graceful degradation
✅ Multiple AI service layers for reliability

3. File Processing

⚠️ Limited image processing capabilities
⚠️ Complex document parsing may need refinement
✅ PDF and text processing working well

🚀 Quick Start

Prerequisites

Node.js 18+
MongoDB Atlas account
Google Cloud Platform account
YouTube Data API key (optional)

Installation

Clone the repository

git clone <repository-url>
cd ai-career-companion

Install frontend dependencies
```
npm install
```
Install backend dependencies
```
cd project/server
npm install
cd ../..
```

Set up Google Cloud credentials

# 1. Create a Google Cloud project and enable Vertex AI API
# 2. Create a service account with Vertex AI permissions
# 3. Download the service account key JSON file
# 4. Place it in the project/server directory

Configure environment variables

# Create .env file in server directory
cp project/server/.env.example project/server/.env

# Edit .env file and add your credentials:
# GOOGLE_APPLICATION_CREDENTIALS=./your-service-account-key.json
# MONGODB_URI=your_mongodb_connection_string
# YOUTUBE_API_KEY=your_youtube_api_key (optional)

Start the application

# Terminal 1: Start backend
cd project/server
npm start

# Terminal 2: Start frontend
cd project
npm run dev

Access the application
- Frontend: http://localhost:5173
- Backend API: http://localhost:3001

🔧 API Endpoints

Resume Analysis

POST /upload_resume - Upload and analyze resume
POST /auto-detect-role - Auto-detect job role from resume
GET /analysis/:id - Get analysis by ID

Learning Roadmaps

POST /generate-learning-roadmap - Generate AI-powered roadmap
POST /generate-skill-roadmap - Generate skill-specific roadmap
GET /roadmap/:id - Get roadmap by ID
GET /user/:userId/roadmaps - Get user's roadmaps

Career Simulations

POST /start-career-simulation - Start new simulation
POST /generate-simulations - Generate multiple simulations
GET /simulation/:id - Get simulation by ID
PUT /simulation/:id/progress - Update simulation progress

AI Mentor

POST /mentor - Chat with AI mentor
GET /conversations/:userId - Get conversation history
GET /user/:userId/history - Get user interaction history

YouTube Integration

GET /test-youtube/:query - Test YouTube video search

🗄️ Database Schema

Analysis Model

{
  userId: String,
  targetRole: String,
  matchScore: Number,
  skillsPresent: [String],
  skillsMissing: [String],
  recommendations: [String],
  createdAt: Date
}

Roadmap Model

{
  userId: String,
  role: String,
  roadmap: {
    stage_1_critical_gaps: [SkillGap],
    stage_2_important_gaps: [SkillGap],
    stage_3_nice_to_have: [SkillGap],
    learning_resources: Object,
    success_metrics: [String]
  },
  createdAt: Date
}

Simulation Model

{
  userId: String,
  title: String,
  description: String,
  scenarios: [Object],
  progress: Number,
  completed: Boolean,
  createdAt: Date
}

🎨 UI/UX Features

Modern Dashboard

Responsive design with Tailwind CSS
Interactive components with smooth animations
Progress tracking with visual indicators
Badge system for gamification

Learning Experience

Modal-based learning with video integration
Tabbed interfaces for organized content
Progress bars and completion tracking
Interactive skill assessments

AI Chat Interface

Real-time messaging with typing indicators
Message history with search functionality
Context-aware responses from AI mentor
Badge notifications for achievements

🔍 AI Implementation Details

RAG (Retrieval-Augmented Generation)

Vector embeddings for semantic search
Context retrieval from job descriptions
Enhanced responses with relevant information
Fallback mechanisms for reliability

Gemini Integration

Gemini 2.5 Flash for fast responses
Structured prompts for consistent output
Error handling with graceful degradation
Multiple service layers for redundancy

Vector Search

Text embeddings using Google Cloud
Semantic similarity for job matching
Context-aware recommendations
Performance optimization

🧪 Testing & Quality

Error Handling

Comprehensive try-catch blocks throughout the application
Graceful degradation when services are unavailable
User-friendly error messages with actionable guidance
Logging system for debugging and monitoring

Performance

Lazy loading for large datasets
Efficient database queries with proper indexing
Caching mechanisms for frequently accessed data
Optimized file processing with streaming

🚧 Known Limitations

Current Issues

YouTube API Rate Limiting - May hit limits with high usage
File Processing - Complex documents may need refinement
AI Response Consistency - Occasional formatting variations
Mobile Responsiveness - Some components need mobile optimization

Future Improvements

Real-time Collaboration - Multi-user features
Advanced Analytics - Detailed progress tracking
Integration APIs - LinkedIn, GitHub, etc.
Mobile App - Native mobile application

🤝 Contributing

Development Setup

Fork the repository
Create a feature branch
Make your changes
Test thoroughly
Submit a pull request

Code Standards

ESLint configuration for code quality
TypeScript for type safety
Conventional commits for clear history
Comprehensive error handling

📄 License

🏆 Hackathon Submission

Innovation Highlights

Advanced RAG Implementation with vector search
Multi-modal AI Integration (text, documents, videos)
Comprehensive Career Development platform
Real-world Application with practical use cases

Technical Achievements

Scalable Architecture with microservices approach
Robust Error Handling with fallback mechanisms
Modern Tech Stack with best practices
Performance Optimization for production readiness

Demo Instructions

Upload a resume to see AI analysis
Generate a personalized learning roadmap
Chat with the AI mentor for career advice
Try career simulations for skill practice
Explore the badge system and progress tracking

📞 Support

For questions or issues:

GitHub Issues - Report bugs and feature requests
Documentation - Comprehensive API and setup guides
Demo Video - Watch the application in action

Built with ❤️ for Google Cloud Gen AI Exchange Hackathon 2025

Name		Name	Last commit message	Last commit date
Latest commit History 65 Commits
.vscode		.vscode
project		project
.gitignore		.gitignore
README.md		README.md
lt.log		lt.log
netlify.toml		netlify.toml
optimize_payload.json		optimize_payload.json

sujithputta02/RunaGen-AI-Prototype

Folders and files

Latest commit

History

Repository files navigation