Environment Setup

To get started with any of the projects, follow these steps:

1. Create a virtual environment:

   python3.9 -m venv venv
   source venv/bin/activate
   

2. Install common dependencies: All necessary libraries for all projects are consolidated in the root requirements.txt:

   pip install -r requirements.txt
   

3. Set up API Keys: Many projects require API keys for Cerebras (for langchain-cerebras) and/or Tavily (for search tools). Set them as environment variables:

   export api_key=your_api_key
   export TAVILY_API_KEY=your_tavily_api_key # Only if the project uses search tools
   

Directory Structure

langgraph-tutorial/
  GEMINI.md
  branding.json
  README.md
  requirements.txt # Consolidated requirements for all projects
  project01/
    README.md
    main.py
    # project-specific requirements.txt (now optional, but kept for individual project clarity)
  ...
  project20/
    README.md
    main.py
    # project-specific requirements.txt (now optional, but kept for individual project clarity)

Coding & Documentation Guidelines

All code adheres to PEP-8, includes type hints, and follows industry-standard LangChain/LangGraph conventions. Each main.py file is kept concise (≤ 20 lines) with clear inline comments for readability.

Each project folder includes:

• main.py: The core LangGraph agent implementation.
• README.md: Outlines the project's purpose, setup & usage, business value, and extension ideas.

Project List: Your Path to LangGraph Mastery

Here's a brief overview of the 20 projects, designed to progressively build your skills in AI agent development:

1. Basic Chat Agent: Simple state graph for foundational understanding.
2. Tools Integration: Fetch live data via function nodes, enhancing agent capabilities.
3. Memory-Enabled Agent: Intermediate state tracking for contextual conversations.
4. Human-in-the-Loop: Approval node using interrupt for critical oversight.
5. Conditional Edges: Branches based on numeric thresholds for dynamic workflows.
6. Deferred Nodes: Map-reduce style aggregation for efficient data processing.
7. Node-Level Caching: Reduce redundant model calls, optimizing performance and cost.
8. Finance Assistant: Graph agent querying stock prices for real-time financial insights.
9. Marketing Generator: Cycle to refine slogans, boosting creative content generation.
10. HR Policy Helper: Stateful Q&A from document, streamlining HR support.
11. Multi-Agent Workflow: Agents coordinate via graph for complex task automation.
12. Email Parser + Responder: Tool invocation graph for automated communication.
13. Interview Coach: Retrieve, assess candidate responses, aiding recruitment.
14. Budget Planner: CSV input → summary graph for financial management.
15. Resume/Job-Match Bot: Multi-node compare graph for talent acquisition.
16. QA over Knowledge Base: Integrate vector DB for intelligent information retrieval.
17. Log Analyzer: Parse logs, summarize errors for system health monitoring.
18. Report Generator: Fetch and compile insights for automated reporting.
19. Command-Line Tool Agent: Shell command node for system interaction.
20. Capstone Unified Assistant: Finance + HR + Marketing orchestration – a powerful, integrated AI solution.

Business Value & Pro Tips

Each project's README.md includes a dedicated "Business Value" section, highlighting how the agent supports real-world domains. We also suggest future enhancements, such as GUI integration, database integration, or deployment to platforms like LangGraph Cloud, to inspire further development.

How to Run

To run any specific project, navigate to its directory (e.g., langgraph-tutorial/project01), ensure your environment variables for API keys are set, and then execute python main.py.

Start your journey to LangGraph mastery with HERE AND NOW AI today!