Author: Karun Swaroop
A multi-agent AI system that generates personalized AI learning paths based on your current skills and career goals. Built with a modern React frontend and FastAPI backend, powered by Claude/OpenAI and RAG (Retrieval Augmented Generation).
- Personalized Learning Paths - AI analyzes your profile and target job to create custom 5-chapter learning journeys
- 153+ Skills Ontology - Comprehensive GenAI skills taxonomy across 18 domains
- Multi-Agent Architecture - Specialized AI agents for profile analysis, JD parsing, gap analysis, and content generation
- 12 Example Profiles - Pre-built personas across industries (Marketing, Healthcare, Finance, Legal, Education, etc.)
- Progress Tracking - Track chapter completion and skill development
- Beautiful UI - Modern, responsive interface with smooth animations
- Frontend: https://ai-pathway-frontend-761055070184.us-central1.run.app
- Backend API: https://ai-pathway-backend-761055070184.us-central1.run.app
- Python 3.11+
- Node.js 18+
- Anthropic API key (or OpenAI)
cd backend
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
# Configure environment
cp .env.example .env
# Edit .env with your API keys
# Run server
python -m uvicorn app.main:app --reload --port 8000cd frontend
npm install
npm run devVisit http://localhost:5173
┌─────────────────────────────────────────────────────────────────┐
│ FRONTEND │
│ React + TypeScript + Tailwind CSS │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │ Home │ │ Profiles │ │ Analysis │ │ Learning │ │
│ │ Page │ │ Page │ │ Page │ │ Path │ │
│ └──────────┘ └──────────┘ └──────────┘ └──────────┘ │
└─────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────────┐
│ BACKEND (FastAPI) │
└─────────────────────────────────────────────────────────────────┘
│
┌─────────────────────┼─────────────────────┐
▼ ▼ ▼
┌───────────────┐ ┌───────────────┐ ┌───────────────┐
│ PROFILE │ │ JD PARSER │ │ GAP ANALYSIS │
│ ANALYZER │ │ AGENT │ │ AGENT │
└───────────────┘ └───────────────┘ └───────────────┘
│
▼
┌───────────────┐
│ PATH │
│ GENERATOR │
└───────────────┘
│
┌───────────────┼───────────────┐
▼ ▼ ▼
┌──────────┐ ┌──────────┐ ┌──────────┐
│ LLM │ │ RAG │ │ Database │
│ (Claude) │ │(ChromaDB)│ │ (SQLite) │
└──────────┘ └──────────┘ └──────────┘
| Layer | Technology |
|---|---|
| Frontend | React 18, TypeScript, Vite, Tailwind CSS, TanStack Query |
| Backend | Python, FastAPI, Pydantic |
| LLM | Claude 3.5 Sonnet (or OpenAI GPT-4) |
| Vector DB | ChromaDB |
| Database | SQLite (dev) / PostgreSQL (prod) |
| Embeddings | sentence-transformers (all-MiniLM-L6-v2) |
| Deployment | Google Cloud Run, Artifact Registry |
ai-pathway/
├── backend/
│ ├── app/
│ │ ├── main.py # FastAPI application
│ │ ├── agents/ # Multi-agent system
│ │ │ ├── orchestrator.py # Coordinates all agents
│ │ │ ├── profile_analyzer.py
│ │ │ ├── jd_parser.py
│ │ │ ├── gap_analyzer.py
│ │ │ └── path_generator.py
│ │ ├── services/
│ │ │ ├── llm/ # LLM abstraction layer
│ │ │ └── rag/ # RAG with ChromaDB
│ │ ├── models/ # SQLAlchemy models
│ │ ├── schemas/ # Pydantic schemas
│ │ └── data/
│ │ ├── ontology.json # 153+ skills ontology
│ │ └── profiles/ # 12 example profiles
│ └── requirements.txt
│
├── frontend/
│ ├── src/
│ │ ├── components/
│ │ │ └── ontology/ # Domain grid visualization
│ │ ├── pages/
│ │ │ ├── HomePage.tsx
│ │ │ ├── ProfileSelectionPage.tsx
│ │ │ ├── AnalysisPage.tsx
│ │ │ ├── LearningPathPage.tsx
│ │ │ └── DashboardPage.tsx
│ │ ├── services/api.ts # API client
│ │ └── types/index.ts # TypeScript types
│ ├── package.json
│ └── tailwind.config.js
│
├── docs/
│ ├── SYSTEM_DESIGN.md # Architecture details
│ ├── FRONTEND_UI.md # UI/UX documentation
│ ├── API_REFERENCE.md # API documentation
│ └── ONTOLOGIES.md # Skills ontology docs
│
└── deploy.sh # GCP deployment script
| # | Name | Current Role | Target Role | Industry |
|---|---|---|---|---|
| 1 | Alex Rivera | Marketing Manager | AI Product Manager | Consumer Goods |
| 2 | Bethany Chen | High School Teacher | AI Learning Designer | EdTech |
| 3 | Charles Patel | Legal Associate | AI Governance Scientist | Legal/RegTech |
| 4 | Dana Morales | Healthcare Data Analyst | Healthcare Data Scientist | Healthcare |
| 5 | Elena Brooks | Financial Analyst | AI Risk Analyst | Finance |
| 6 | Frank Nguyen | HR Manager | People Analytics Lead | HR |
| 7 | Grace Williams | VP Marketing | AI Marketing Strategist | Retail |
| 8 | Hank Thompson | VP Operations | AI Transformation Leader | Manufacturing |
| 9 | Irene Shah | CIO | Enterprise AI Leader | Insurance |
| 10 | John Miller | Software Developer | ML Engineer | Technology |
| 11 | Kelly Johnson | Content Specialist | Prompt Engineer | Marketing |
| 12 | Kevin Park | Automation Engineer | Robotics AI Engineer | Manufacturing |
The system uses a comprehensive skills taxonomy:
- 153+ Skills across 18 domains
- 5-Level Proficiency Scale: L0 (None) → L4 (Expert)
- 7 Layers: Foundation, Theory, Application, Tools, Tech Prerequisites, Domain, Soft/Strategy
- Prerequisite Mapping: Ensures logical learning progression
| Layer | Domains |
|---|---|
| Foundation | Digital Literacy, Critical Thinking |
| Theory | AI Foundations |
| Application | Prompting & HITL, RAG Systems, Agents, Model Adaptation, Multimodal AI, Evaluation, Safety & Security, LLMOps |
| Tools | Tools & Frameworks |
| Tech Prerequisites | Tech Prerequisites |
| Domain | Governance, Domain Apps |
| Soft/Strategy | Product & UX, Communication, Learning & Adaptation |
| Method | Endpoint | Description |
|---|---|---|
| GET | /api/profiles |
List all example profiles |
| GET | /api/profiles/{id} |
Get profile details |
| POST | /api/analysis/full |
Run full analysis pipeline |
| GET | /api/paths/{id} |
Get learning path |
| PUT | /api/progress/{path_id} |
Update chapter progress |
| GET | /api/ontology/ |
Get skills ontology |
See API Reference for complete documentation.
# LLM Configuration
LLM_PROVIDER=claude # claude or openai
ANTHROPIC_API_KEY=your-key
OPENAI_API_KEY=your-key
# Models
CLAUDE_MODEL=claude-3-5-sonnet-20241022
OPENAI_MODEL=gpt-4-turbo-preview
# Database
DATABASE_URL=sqlite+aiosqlite:///./ai_pathway.db
# ChromaDB
CHROMA_PERSIST_DIRECTORY=./chroma_db
# Embeddings
EMBEDDING_MODEL=all-MiniLM-L6-v2
# CORS (for deployment)
CORS_ORIGINS=["http://localhost:5173"]The application is deployed on Google Cloud Run. See deploy.sh for the deployment script.
# Build and deploy backend
cd backend
docker build --platform linux/amd64 -t us-central1-docker.pkg.dev/ai-pathway-486221/ai-pathway/backend:v1 .
docker push us-central1-docker.pkg.dev/ai-pathway-486221/ai-pathway/backend:v1
gcloud run deploy ai-pathway-backend --image ... --region us-central1
# Build and deploy frontend
cd frontend
docker build --platform linux/amd64 -t us-central1-docker.pkg.dev/ai-pathway-486221/ai-pathway/frontend:v1 .
docker push us-central1-docker.pkg.dev/ai-pathway-486221/ai-pathway/frontend:v1
gcloud run deploy ai-pathway-frontend --image ... --region us-central1Note: When building on Apple Silicon, always use --platform linux/amd64 for Cloud Run compatibility.
- System Design - Architecture and design decisions
- Frontend UI - UI/UX documentation and component guide
- API Reference - Complete API documentation
- Ontologies - Skills ontology documentation
- Fork the repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Open a Pull Request
MIT License - see LICENSE for details.
- Built with Claude by Anthropic
- Skills ontology developed by Colaberry
- UI inspired by modern design systems