New course with Google Cloud: LLMOps

Hi, I'm delighted to announce this short course on llmops built in partnership with Google Cloud and taught by Irvin Hausenher. Machine Learning operations, or MLOps, is the practice of managing how you build and then deploy models in production. In this course you learn how to do this for large language models. LLMOPS is an exciting emerging field and is also making the life of developers much easier through automation to make much faster. A lot of the otherwise manual steps of building an LLM application, such as downloading and trying out different large language models like do you use Palm or Llama or something else and what works best, or trying out different prompts and evaluating performance rigorously and then deploying, monitoring model performance and even figuring out what to do if the LM provider updates their model that you had already built on. Because application building using LMS is still a relatively new field, LMOPS2 is a new field with emerging best practices and in this short course you learn this cutting edge of LMOPS thinking. I'm delighted to introduce the instructor Erwin Hausenha, developer Advocate for Generative AI at Google Cloud. Erwin has helped many developers with with the two chain and processes for building with LLMs and I look forward to sharing his deep practical experience with us here.

Thanks Andrew. It has been gratifying to see how good llmops thinking has helped many developers to do more fun work and less annoying repetitive work. I find that a good pipeline actually makes building more fun. In this course you'll learn to build a machine learning operations workflow that processes data, orchestrates a supervised fine tuning job and deploys the model as an API. We will be covering the latest concepts of operations of large language models, so.

Come learn to avoid your LLM oops with your LLM ops. I hope you enjoy the course.