WHAT DOES MLOps LOOK LIKE FOR LARGE LANGUAGE MODELS (LLMs) IN PRODUCTION?

WHAT DOES MLOps LOOK LIKE FOR LARGE LANGUAGE MODELS (LLMs) IN PRODUCTION?
In the BC (Before- ChatGPT!) era, MLOps primarily encompassed the processes related to model serving, inference and monitoring. During that time, considerable effort was dedicated to data collection and model training, with MLOps serving as a means to operationalise and manage those models effectively. However, the landscape has shifted post-BC and the traditional practices of model training and fine-tuning have become less common due to extended context window size. While all of these core processes still remains crucial, in this talk we deep dive into emerging dynamics of operatisaling LLMs at scale in addition to invaluable insights gleaned from the real-world industrial applications.
Arezou Soltani
Atlassian, Senior Machine Learning Engineer
Arezou is a Senior Machine Learning Engineer at Atlassian and has contributed to the development of the conversation engine powering Atlassian Intelligence, the Virtual Assistant for Jira Service Management. She was awarded a PhD in Computer Science from RMIT University, and is the recipient of VentureBeat’s Women in AI Leadership Award 2021.
Arezou's industrial experience relates to the field of MLOps aiming at productionalising large-scale machine learning models. She believes her academic roots have laid the foundation for her transition to an industrial career, where she can push the limits of traditional software engineering toward innovative machine learning products.
Find out more about DataEngBytes at our website:
dataengbytes.com/