LangChain Training: Build LLM Powered Applications

Description:

Ready to build cutting-edge applications powered by Large Language Models (LLMs)? Using LangChain is an open-source framework for building applications using Python. LangChain provides modular and user-friendly abstractions that can interface with any LLM – OpenAI, Hugging face, Bard, PaLM, etc. Using LangChain, you can develop innovative LLM solutions for wide variety of problems: Text Summarization, Answering Question using sources, analyzing structured data and much more. In this course, you will learn how to use various components in LangChain such as Model input-output, data connection, chains, agents and memory, you will also learn how to integrate vector sources such as FAISS and ChromaDB.

Duration: 1 Day

Course Code: BDT303

Learning Objectives:

After this course, you will be able to:

• Develop an understanding of what LangChain is and what are the different components of LangChain
• Explore model inputs and outputs when using prompts
• Understand how to connect the models to data sources, explore the document integrators, text embeddings and Vector Stores
• Learn how to build a chain that connects one call to another to build a robust application
• Understand how to leverage memory to keep track of message interactions with the model
• Learn how to build agents that will use different LangChain components to build applications

• Must have some python programming experience.

• This course is designed for software engineers, backend developers, full stack developers or business analysts who want to build Generative AI based applications with LangChain – no machine learning experience required.

• This course is designed for software engineers, backend developers, full stack developers or business analysts who want to build Generative AI based applications with LangChain – no machine learning experience required.

1. Introduction to LangChain

• Understanding what is LangChain?
• Installing LangChain software
• Creating a simple prompt
• Hands-on: Create a simple prompt

2. Using Model Inputs and Outputs

• Introduction to Model: Inputs and Outputs
• Using an LLM with LangChain
• Using Prompt Templates with LangChain – zero, one and few shot prompts
• Hands-on lab with Model inputs and outputs

3. Understanding Data Connections

• Data connection introduction
• Document Loaders – Integrations
• Document Transformers
• Text Embeddings and leveraging Vector Stores
• Hands-on lab with these topics

4. Using Chains

• LLM Chain Object
• Understanding different LLM Chains
• Sequential and LLM Router Chain
• Hands-on lab with these topics

5. Using Memory for prior chats

• Introduction to Memory
• Different types of memories and memory buffers
• Hands-on lab with Memory

6. Agents

• Introduction to Agents
• Agents use of ReACT framework
• Explore different Agent Tools
• Hands-on lab with these topics

Training material provided: Yes (Digital format)

Hands-on Lab: Instructions will be provided to install Jupyter notebook and other required python libraries. Students can opt to use ‘Google Colaboratory’ if they do not want to install these tools