LangChain Runnable Chains

A collection of practical LangChain workflow implementations built using Groq LLMs and LangChain's runnable architecture.

This project explores different chaining patterns such as sequential pipelines, branching logic, parallel execution, and structured output parsing using real-world examples.

---

What's Included

Simple Chain

Basic prompt → model → parser workflow for generating AI responses.

Sequential Chain

Multi-step workflow where the output of one chain becomes the input for another.

Parallel Chain

Runs multiple tasks simultaneously and combines the results into a single output.

Branching Chain

Uses conditional routing to generate different responses based on sentiment classification.

---

Tech Stack

• Python
• LangChain
• Groq API
• Pydantic
• Runnable Chains
• Prompt Templates

---

Project Structure

langchain-runnable-chains/
│
├── chains/
│   ├── simple_chain.py
│   ├── sequential_chain.py
│   ├── parallel_chain.py
│   └── branching_chain.py
│
├── screenshots/
│
├── .env.example
├── .gitignore
├── requirements.txt
└── README.md

---

Setup

Clone the repository

git clone https://github.com/your-username/langchain-runnable-chains.git

Move into the project directory

cd langchain-runnable-chains

Install dependencies

pip install -r requirements.txt

---

Environment Variables

Create a .env file and add:

GROQ_API_KEY=your_groq_api_key

---

Running the Examples

Example:

python chains/simple_chain.py

You can similarly run:

• sequential_chain.py
• parallel_chain.py
• branching_chain.py

---

Concepts Covered

• Runnable chains
• Prompt pipelines
• Output parsers
• Structured responses
• Parallel execution
• Conditional branching
• Chain visualization
• LLM workflow orchestration

---

Screenshots

Add terminal outputs or chain graphs inside the screenshots/ folder and reference them here.

Example:

![Chain Output](screenshots/demo.png)

---

Author

Vrundali Rahangdale