LLMOps in Practice: A Deep Dive
4h 27mIntermediate2024-12-18
Authors
Laurence Moroney
Course details
As LLM-based apps proliferate, many function as thin veneers around more robust models such as GPT. In this course, instructor Laurence Moroney demonstrates the basics of building your own LLM-powered app and controlling how it works from the back end. Get an introduction to some of the basic use cases of LLMs in application development. Basic knowledge of Python, Node.js, and LLMs is required. By the end of this course, you’ll be prepared to create an ops ecosystem whose core functionality is based on an LLM.
Learning objectives
Build a basic LLM app and understand ongoing operations.
Understand the basic concepts of building an application that uses a hosted model.
Create a web application that wraps an LLM using an API.
Use RAG to extend your application.
Create a vector database using PineCone or VectorDB.
Learning objectives
Build a basic LLM app and understand ongoing operations.
Understand the basic concepts of building an application that uses a hosted model.
Create a web application that wraps an LLM using an API.
Use RAG to extend your application.
Create a vector database using PineCone or VectorDB.
Skills covered
Natural Language Processing (NLP)Generative AIPythonArtificial Intelligence (AI)Open SourceOne-Off
Concepts
0. Introduction
- 01 - A deep dive into LLM operations
- 02 - What are LLMs
- 03 - What are transformers
- 04 - What is LLMOps
1. Build a Basic LLM App
- 05 - Prompting
- 06 - Advanced prompting
- 07 - Hosting an app
- 08 - Create a chatbot
- 09 - LLM exercise
- 10 - Adding logging to your server
2. First Steps in Ops
- 11 - Coding for logging
- 12 - Exploring the logging system
- 13 - RLHF and user feedback
- 14 - Challenge - Implementing RLHF and user feedback
- 15 - Demonstrating the ops project completed
- 16 - Solution - Completing an ops project
- 17 - Demonstrating the code for the ops
3. BYOD with RAG
- 18 - Retrieval augmented generation (RAG)
- 19 - Installing and setting up a VectorDB
- 20 - Create a VectorDB
- 21 - BYOD to a VectorDB
- 22 - VectorDB - Hands-on use case
- 23 - Querying the VectorDB
- 24 - Demonstration - Querying the VectorDB
- 25 - Extending your app with RAG
- 26 - RAG - Showing it in action
- 27 - Challenge - Complete RAG application
- 28 - Solution - Complete RAG application
4. RAG and Ops
- 29 - Extending RAG with ops
- 30 - Logging
- 31 - Hands-on logging
- 32 - Metrics
- 33 - Hands-on metrics
- 34 - Version management
- 35 - Hands-on version management
- 36 - RAG Ops - Updating the data
- 37 - Hands-on RAG ops
- 38 - RAG in action - Exercise
Conclusion
- 39 - Continue your LLMOps learning journey