This repository contains code for building a conversational AI assistant using the LangChain library and large language models like GPT-3. It accompanies the Udemy course "LangChain - Develop LLM powered applications with LangChain".
Overview This course covers creating chatbots and intelligent assistants by leveraging large language models in an effective way using the Python LangChain library. Key topics include:
Using LangChain's tools like reasoning engines and action selectors to structure conversations Writing effective prompts to get the most out of large models like GPT-3 Techniques like chain of thought prompting and few shot learning Building a "react" agent that can select the right tool/action based on the prompt Wrapping text processing functions and more into LangChain tools Using models like GPT-3 and Pinecone for reasoning and embeddings Building real-world applications like a LinkedIn scraper, Twitter analyzer, documentation reader and more Prompt engineering strategies to optimize assistant performance Usage The assistant.py file contains the core assistant code. Configure your OpenAI API key and other credentials in .env to run it.
The tools/ folder contains various utility functions wrapped as LangChain tools.
See the other folders for code for specific applications developed in the course.
Installation
pip install -r requirements.txt