Build Conversational Agents with Vector DBs - LangChain #9
We've seen in previous chapters how powerful retrieval augmentation (vector databases) and conversational agents (chatbots) can be. They become even more impressive when we begin using them together. Conversational agents can struggle with data freshness, knowledge about specific domains, or accessing internal documentation. By coupling agents with retrieval augmentation tools, we no longer have these problems. On the other side, using "naive" retrieval augmentation without the use of an agent means we will retrieve contexts with every query. Again, this isn't always ideal as not every query requires access to external knowledge. Merging these methods gives us the best of both worlds. In this video, we'll learn how to do using the Pinecone vector database and OpenAI embedding and gpt-3.5-turbo completion models. 📌 Code notebook: https://github.com/pinecone-io/examples/blob/master/learn/generation/langchain/handbook/08-langchain-retrieval-agent.ipynb 👋🏼 NLP + LLM Consulting: https://aurelio.ai 🎙️ Support me on Patreon: https://patreon.com/JamesBriggs 👾 Discord: https://discord.gg/c5QtDB9RAP 00:00 LangChain Agents with Vector DBs 01:27 Code Setup and Data Prep 03:14 Vector DB Pipeline Setup 05:35 Indexing with OpenAI and Pinecone 07:53 Querying via LangChain 09:33 Building the Retrieval Augmented Chatbot 13:52 Using the Conversational Agent Chatbot 17:17 Real-world Usage of this Method #artificialintelligence #langchain #openai #chatgpt #nlp #deeplearning
