Overview

This course features Coursera Coach! A smarter way to learn with interactive, real-time conversations that help you test your knowledge, challenge assumptions, and deepen your understanding as you progress through the course. In this comprehensive course, you'll learn how to build and fine-tune large language models (LLMs) for real-world applications. Starting with fundamental concepts, you'll progress through hands-on projects that focus on document-based retrieval-augmented generation (RAG) systems, LangChain integration, and fine-tuning techniques. You'll gain the skills to build custom applications like a PDF RAG system, a voice assistant, and a YouTube video summarizer, with a focus on optimizing the retrieval and generation of content. With a blend of theoretical lessons and practical exercises, this course ensures you master both building and fine-tuning LLMs for various AI-driven tasks. You'll also dive deep into advanced fine-tuning methods like LoRA (Low-Rank Adaptation), learning to fine-tune models efficiently with minimal computational resources. Throughout the course, you'll implement real-world projects that integrate sophisticated LLM functionalities into usable applications. By the end of the course, you’ll be capable of deploying and fine-tuning LLMs for personalized tasks, giving you the tools to tackle complex AI challenges in your own projects. This course is designed for intermediate to advanced learners with prior programming experience. It’s perfect for those who want to deepen their understanding of LLMs and apply them to solve industry-specific problems. No prior experience with fine-tuning is required, though knowledge of Python and machine learning basics will be beneficial. By the end of the course, you will be able to build, fine-tune, and deploy LLM applications, including RAG systems, voice assistants, and specialized chatbots, using advanced techniques such as LoRA fine-tuning.

Syllabus

Hands-On: PDF RAG System with Text Chunking

In this module, we will walk you through building a PDF RAG system using text chunking and overlap strategies. You’ll explore the system architecture, set up the necessary components, and test your system to ensure it retrieves information effectively, processing large documents in manageable chunks.

LangChain Fundamentals and Workflow Integration

In this module, we will introduce you to LangChain, an essential framework for building robust LLM applications. You’ll dive deep into LangChain's structure, set up powerful components like chat models and text processors, and build hands-on systems using LangChain’s dynamic tools for data management and retrieval.

Hands-On: Building LLM Applications with LangChain

In this module, we will guide you through building real-world LLM applications using LangChain. You’ll implement various systems, such as a news summarizer, YouTube video summarizer, and voice assistant RAG system, using LangChain's powerful capabilities to process and generate responses from text and voice inputs.

Fine-Tuning LLMs

In this module, we will dive into fine-tuning techniques, teaching you how to customize LLMs for specific applications. You’ll go through the entire fine-tuning process, from selecting an appropriate technique to applying it to your own dataset and testing the resulting model with real-world queries.

LoRA-Based Fine-Tuning and Deployment

In this module, we will focus on LoRA, a powerful technique for efficiently fine-tuning LLMs. You will learn the theory behind LoRA, implement it using PEFT strategies, and then deploy your fine-tuned models as API services for easy integration and testing.

Wrap-Up and Next Steps

In this module, we will summarize the major learnings from the course, helping you solidify your understanding of LLM application development and fine-tuning. You’ll also get recommendations for next steps to continue advancing in the field of AI.