Getting Started¶

Welcome to the Agentic AI Nano-Degree! This guide will help you set up your environment and begin your learning journey.

Quick Start¶

Choose Your Module: Start with Agent Frameworks or RAG Architecture
Select Learning Path: Observer, Participant, or Implementer based on your time and depth needs
Set Up Environment: Follow the setup instructions below
Begin Learning: Start with Session 0 of your chosen module

Prerequisites¶

Required Knowledge¶

Python Programming: Intermediate level (functions, classes, modules)
Command Line: Basic terminal/command prompt usage
Git: Basic version control operations
APIs: Understanding of REST APIs and HTTP

Recommended Background¶

Machine Learning: Basic understanding of ML concepts
Databases: Familiarity with SQL and NoSQL databases
Cloud Platforms: Exposure to AWS, GCP, or Azure
Docker: Basic containerization concepts

Hardware Requirements¶

RAM: Minimum 8GB, recommended 16GB+
Storage: 10GB free space for code examples and models
Internet: Stable connection for API calls and model downloads
OS: Windows 10+, macOS 10.15+, or Linux (Ubuntu 18.04+)

Environment Setup¶

1. Python Environment¶

We recommend using Python 3.9 or higher with a virtual environment:

# Create virtual environment
python -m venv agentic-ai
source agentic-ai/bin/activate  # On Windows: agentic-ai\Scripts\activate

# Upgrade pip
pip install --upgrade pip

2. Clone Repository¶

git clone https://github.com/yourusername/nano-degree.git
cd nano-degree

3. Install Dependencies¶

# Install core requirements
pip install -r requirements.txt

# For specific modules, install additional dependencies:
# Agent Frameworks
pip install -r 01_frameworks/requirements.txt

# RAG Architecture  
pip install -r 02_rag/requirements.txt

4. API Keys Setup¶

Create a .env file in the project root:

# OpenAI (required for most examples)
OPENAI_API_KEY=your_openai_api_key_here

# Optional: Other LLM providers
ANTHROPIC_API_KEY=your_anthropic_key_here
COHERE_API_KEY=your_cohere_key_here

# Vector Database APIs (choose based on your preference)
PINECONE_API_KEY=your_pinecone_key_here
WEAVIATE_URL=your_weaviate_instance_url
CHROMA_PERSIST_DIRECTORY=./chroma_db

# Other services
GOOGLE_API_KEY=your_google_key_here
HF_TOKEN=your_huggingface_token_here

5. Verify Installation¶

# Test core functionality
python -c "import openai, langchain, chromadb; print('Setup complete!')"

# Run a simple test
cd 01_frameworks/src/session1
python test_setup.py

Learning Paths Guide¶

👀 Observer Path (30-80 min/session)¶

Best for: Quick understanding, decision makers, busy schedules

Activities: - Read core concepts and architectural overviews - Examine code examples and patterns - Understand design decisions and trade-offs - Review use cases and applications

Setup: Minimal - just read the materials

🙋‍♂️ Participant Path (50-110 min/session)¶

Best for: Hands-on learners, developers, practical understanding

Activities: - Follow guided implementations step-by-step - Run provided code examples - Complete exercises and mini-projects - Analyze and modify existing implementations

Setup: Full environment with API keys

🛠️ Implementer Path (120-260 min/session)¶

Best for: Deep expertise, architects, production focus

Activities: - Build complete systems from scratch - Implement custom components and optimizations - Explore advanced patterns and architectures - Create production-ready solutions

Setup: Full environment plus additional tools (Docker, cloud access)

Project Structure¶

nano-degree/
├── 01_frameworks/          # Agent Frameworks Module
│   ├── Session0_Introduction/
│   ├── Session1_Bare_Metal/
│   ├── ...
│   └── src/                # Source code examples
├── 02_rag/                 # RAG Architecture Module  
│   ├── Session0_Introduction/
│   ├── Session1_Basic_RAG/
│   ├── ...
│   └── src/                # RAG implementations
├── docs/                   # Documentation and guides
├── tests/                  # Test suites
├── requirements.txt        # Core dependencies
└── README.md              # Project overview

Study Recommendations¶

Time Management¶

Consistent Schedule: Set aside dedicated learning time
Session Completion: Finish sessions completely before moving on
Practice Time: Allow extra time for hands-on practice
Review Period: Revisit complex concepts after a few days

Learning Strategies¶

Start with Foundations: Don't skip early sessions
Practice Immediately: Run code examples as you learn
Take Notes: Document insights and modifications
Join Community: Engage with other learners
Apply Practically: Use concepts in real projects

Common Schedules¶

Intensive (2-3 weeks): - 2 sessions per day (Observer path) - 1 session per day (Participant path)
- 1 session every 2 days (Implementer path)

Regular (4-6 weeks): - 1 session per day (Observer path) - 1 session every 2 days (Participant path) - 2-3 sessions per week (Implementer path)

Extended (8-12 weeks): - 3-4 sessions per week (any path) - More time for practice and projects - Deep exploration of optional modules

Development Tools¶

Recommended IDEs¶

VS Code: Excellent Python support, extensions
PyCharm: Professional Python IDE
Jupyter: Great for experimentation and learning

Useful Extensions/Plugins¶

Python syntax highlighting and linting
Git integration
Docker support
Markdown preview
Code formatting (Black, autopep8)

Command Line Tools¶

# Code formatting
pip install black isort

# Linting
pip install flake8 pylint

# Testing
pip install pytest pytest-cov

# Documentation
pip install mkdocs mkdocs-material

🆘 Support & Community¶

Getting Help¶

Documentation: Check session materials and API docs first
GitHub Issues: Report bugs or unclear instructions
Discussions: Join community discussions for questions
Stack Overflow: Use tags agentic-ai, langchain, rag

Community Resources¶

GitHub Discussions: Course-specific questions and sharing
Discord/Slack: Real-time chat with other learners
Office Hours: Weekly community calls (check schedule)
Study Groups: Form groups with other learners

Troubleshooting¶

Common Issues:

API Rate Limits: Use smaller examples, implement delays
Memory Issues: Close other applications, use smaller models
Import Errors: Check virtual environment activation
Network Issues: Verify API keys and internet connection

Debug Steps: 1. Check error messages carefully 2. Verify environment variables 3. Test with minimal examples 4. Check API status pages 5. Ask for help with specific error details

Progress Tracking¶

Session Completion¶

Module Milestones¶

Session 0-2: Foundations established
Session 3-5: Intermediate concepts mastered
Session 6-8: Advanced patterns implemented
Session 9: Production deployment ready

Portfolio Projects¶

Consider building these throughout your learning:

Personal Assistant Agent (Sessions 1-4)
Document Q&A System (Sessions 1-3 RAG)
Multi-Agent Collaboration (Sessions 7-9)
Enterprise RAG System (Sessions 6-9 RAG)

Certification¶

Upon completion, you can: - Document Progress: Keep a portfolio of implementations - Share Projects: Publish code examples and write-ups - Join Alumni: Access to ongoing community and updates - Contribute: Help improve course materials

Ready to start learning? Choose your path and dive into the world of Agentic AI!

Start with Agent Frameworks → Begin with RAG Architecture →