Large Language Models

This course explores the theoretical concepts, design principles, and practical applications of large language models (LLMs). Students will gain a deep understanding of how LLMs process and generate human language, with a focus on both conceptual understanding and project-based application for real-world tasks. Topics covered include the transformer architecture, self-attention mechanisms, prompt engineering, model fine-tuning, and advanced applications like retrieval and agent systems. This course gives students practical experience implementing and customizing LLMs for applications such as text generation, classification, summarization, semantic search, and question answering, and they will work with both open-source and proprietary tools and frameworks, including LangChain, Hugging Face, and OpenAI. By the end of the course, students will be able to fine-tune pre-trained LLMs, evaluate their performance, and apply them effectively. Prior experience with Python is required; prior experience with deep learning will be beneficial.