# langchain-agents-with-watsonx

**Repository Path**: mirrors_ibm/langchain-agents-with-watsonx

## Basic Information

- **Project Name**: langchain-agents-with-watsonx
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2024-09-13
- **Last Updated**: 2025-08-23

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Use watsonx.ai and LangChain Agents to perform E-commerce Analytics

This asset demonstrates how to use large language models, e.g. mistralai/mistral-large, to create Structured Chat LangChain Agent with memory, where an Agent may perform sequence of actions based on the model reasoning and assures that LLM uses chat memory to provide the most accurate answer on the user question. The Agent can also use various tools such as SQL or RAG retriever to get the relevant information from structured data sources such as SQL tables or unstrustured data sources such as Vectorstore. We have showcased how an e-commerce analysis can be performed with the LLM Agents.

## Steps:
1. Go to [IBM TechZone](https://techzone.ibm.com/collection/tech-zone-certified-base-images/journey-watsonx) and select **watsonx.ai/,governance SaaS** and reserve it. When you make the reservation, ensure you select to **install Db2**.
2. Once the TechZone environment is procured, go to <cloud.ibm.com/resources> and open the Db2. Create tables in the Db2 Database with the ecommerce datasets provided in the [data_asset/](data_asset) directory. Your Db2 should have these tables:
  - customer_purchase_history
  - product_inventory
  - sales_data
3. Go to [watsonx dashboard](https://dataplatform.cloud.ibm.com/) create a sandbox project and import the notebook from [notebook/](notebook).
4. Follow the instructions in the notebook to execute the use case.