The goal of this course is to teach you everything you need to build real-world, production software with large language models. Unlike other LLM-focused courses, our emphasis will not be on theory or on building simple proof-of-concept projects in Jupyter notebooks. We will instead be exploring the operations behind production-grade LLM applications. This includes, but is not limited to experiment management and model versioning, fine-tuning and prompt engineering, deployment, monitoring, and maintenance of LLMs, and AI safety and security (including how to protect against adversarial actors). Along the way, we’ll build several full applications, including a chatbot, an evaluation system, and a clickbait detector. Finally, we’ll look at the future of LLM and LLMOps.