Introduction to AI Orchestration with LangChain and LlamaIndex
1h 28mIntermediate2024-02-16
Authors
M. Joel Dubinko
Course details
Are you ready to dive into the world of AI applications? This course was designed for you. AI orchestration frameworks let you step back from the details of artificial intelligence tools and APIs and instead focus on building more general, effective systems that solve real-world problems. Join instructor M.Joel Dubinko as he explores the business benefits of AI orchestration—faster development, smarter interfaces, lower costs, and more. This course provides an overview of AI fundamentals and key capabilities, like accessing external tools and databases, with a special focus on exploring local models running on your own hardware, alongside or instead of cloud services like those from OpenAI. Every step of the way, Joel offers hands-on demonstrations of two industry-leading frameworks: LangChain and LlamaIndex. By the end of this course, you’ll be prepared to start building chatbots, intelligent agents, and other useful tools, while monitoring for errors and troubleshooting as you go.
Skills covered
LlamaIndexLangChainNatural Language Processing (NLP)Generative AIArtificial Intelligence FoundationsSoftware Development ToolsArtificial Intelligence (AI)Open SourceSoftware DevelopmentOne-Off
Concepts
0. Introduction
- 01 - Building local AI apps with LangChain and LlamaIndex
- 02 - What you should know
- 03 - Setting up your environment for building AI apps
1. Use AI Orchestration to Build Your First App
- 04 - AI orchestration concepts
- 05 - Building an app with the OpenAI API
- 06 - Running local LLMs
- 07 - Your first LangChain app
- 08 - Your first LlamaIndex app
- 09 - Debugging AI apps
2. Combine LLMs and Indexes to Query Local Documents
- 10 - AI over local documents - Retrieval-augmented generation
- 11 - Choosing an embedding
- 12 - RAG with LlamaIndex
- 13 - RAG with LangChain
- 14 - Challenge - Document summarization
- 15 - Solution - Document summarization
3. Assemble Multi-Step AI Workflows with Chaining
- 16 - App concepts for chaining and more complex workflows
- 17 - Getting JSON out of the LLM
- 18 - LLM function calling
- 19 - Challenge - Local LLM task offloading
- 20 - Solution - Local LLM task offloading
4. Let the AI Decide What to Do Next - Building Agents
- 21 - Introduction to the ReAct agent framework
- 22 - Implementing a ReAct agent
- 23 - Challenge - LangChain and LlamaIndex strengths and weaknesses
- 24 - Solution - LangChain and LlamaIndex strengths and weaknesses
Conclusion
- 25 - Next steps for AI app engineers