12-08, 10:00–11:30 (UTC), LLM Track
Chatbots that understand contexts and respond based on past conversations are a "Dream Come True" with state-of-the-art Generative AI models. In this tutorial, I will demonstrate building a chatbot using OpenAI API and LLMs available on HuggingFace. Besides, I will talk about the advantages of using LangChain and the different strategies that can be used to configure your chatbot to yield the best responses. Not only that, the chatbot can also get you the relevant texts (basically the context) from which it derives the answers for transparency, validation and troubleshooting. Python libraries like OpenAI, HuggingFace, LangChain and Streamlit will be used in the majority of the tutorial to build this GenAI-powered chatbot.
Chatbots that understand contexts and respond based on past conversations are a "Dream Come True" with state-of-the-art Generative AI models. In this tutorial, I will demonstrate building a chatbot using OpenAI API and LLMs available on HuggingFace. Besides, I will talk about the advantages of using LangChain and the different strategies that can be used to configure your chatbot to yield the best responses. Not only that, the chatbot can also get you the relevant texts (basically the context) from which it derives the answers for transparency, validation and troubleshooting. Python libraries like OpenAI, HuggingFace, LangChain and Streamlit will be used in the majority of the tutorial to build this GenAI-powered chatbot.
Agenda:
- Introduction to LLMs, GenAI and Chatbots
- Prompt Engineering for ChatBots
- Vector Databases
- Introduction to LangChain and QA strategies
- End-to-end walkthrough of Training a Chatbot
- Plugging in your ChatBot to a Streamlit UI
Prerequisites:
- This is an intermediate-level tutorial with a mix of beginner-friendly introductions to concepts. To make the best of this tutorial and follow along, having a working knowledge of Python, a notebook environment (Jupyter recommended over Collab) and an IDE (like PyCharm, or VSCode) are recommended.
Previous knowledge expected
Data Scientist @ EY (UK & Irleand) | Education Lead @ Women in AI Ireland | Blogger | Speaker | Mentor | <3 NLP