The PYTHON511-PYTHON514 Generative AI and Large Language Models (LLMs) training series introduces practical ways to work with Generative AI and large language models in research and applied settings. Rather than focusing on theory alone, the sessions are designed to help you understand how these tools work, what they can and cannot do, and where they can be applied effectively.
This PYTHON514 course focuses on applying large language models in realistic, end-to-end settings, with a strong emphasis on model evaluation, analysis, and system-level design. Building on the generative workflows introduced in PYTHON512, this module shifts the focus from producing outputs to understanding, assessing, and improving model performance.
Throughout this course, you will learn how to evaluate both pre-trained and self-built generative models using task-specific evaluation metrics and systematic evaluation workflows. You will analyse model behaviour, interpret evaluation results, and explore practical techniques such as sensitivity analysis to better understand why models behave as they do.
You will also be introduced to knowledge-based model applications, including retrieval-augmented generation (RAG), and learn how external knowledge sources can be integrated into generative AI systems to improve reliability and relevance. These concepts will be brought together in a capstone-style, end-to-end pre-trained model application, covering the full workflow from data preparation and modelling to evaluation and output generation.
By the end of this course, you will be able to critically evaluate large language model–based systems, design and assess complete application pipelines, and make informed decisions about model selection, tuning, and deployment in applied NLP and generative AI scenarios.