LangGraph Examples

Environment Setup

Use UV environment manager to run the examples.

# Install uv on MacOS
brew install uv

# Install uv on Windows
curl -LsSf https://astral.sh/uv/install.sh | sh

# Clone the Git repository
git clone <this_repository_url>

# Sync the environment 
uv sync

Configuration

Create a .env file in the root directory and add your GitHub inference credentials:

GITHUB_INFERENCE_ENDPOINT="<github_inference_endpoint>"
GITHUB_TOKEN="<github_token>"
LOG_LEVEL="<log_level>"
OPIK_API_KEY="<opik_api_key>" # Optional for observability
TAVILY_API_KEY="<tavily_api_key>"
DEEPGRAM_API_KEY="<deepgram_api_key>"

The examples are currently configured to use GitHub Models via AzureAIChatCompletionsModel.

If you wish to use a different provider (e.g., OpenAI, Azure OpenAI, or Anthropic), code changes are required.

Example:

1. Open src/graph_examples/doc_generator/doc_gen.py
2. Update the DocGen class initialization to use your preferred LangChain chat model.
3. Update the model parameters to match your provider's available models.

Running the Examples

LangGraph Examples
AI Product Comparision Studio Production Studio powered by LangGraph Agents, Tavily, DuckDuckGo & Deepgram
Sample Output: apple_watch_se3_vs_fitbit_sense.mp3, apple_watch_se3_vs_fitbit_sense.txt From the repository root folder, execute: uv run product_review
Document Generator	Search and Reranking Analysis reranking via FlashRank & ms-marco-MiniLM-L-12-v2 cross-encoder
From the repository root folder, execute: uv run doc_gen	From the repository root folder, execute: uv run rag_search

💡 Heads Up

This is a living repository. Expect regular updates as the design matures, new workflows are added, and existing workflows are refined.