Abstract: Training a great ML model is only the first step. In this demo talk, we will use the open-source MLOps orchestration framework MLRun to deploy your model and make it available to a user in a larger application. MLRun introduces Data Scientists to a simple Python SDK that transforms their code into a production-quality application by abstracting the many layers involved in the MLOps pipeline.
In this talk, we will discuss:
Overview of MLRun capabilities
General model deployment strategies and considerations
Model deployment with MLRun locally and on Kubernetes
Creating an AI app with Huggingface Spaces and MLRun
Bio: Nick is a passionate machine learning, data science, and MLOps enthusiast with experience across multiple domains including fraud detection, natural language processing, computer vision, and data mining. Nick holds a BSc. in Cognitive Science with a specialization in ML and Neural Computation from University of California, San Diego. He is an AWS Certified Solutions Architect, and has earned certifications in Python, Pytorch, Apache Airflow, PySpark and other frameworks. Currently, Nick acts as pre-sales MLOps Engineer at Iguazio, where he specializes in helping enterprises create real-world impact with their data science initiatives, with expertise in deployments on AWS, GCP, and Azure as well as on-premise Kubernetes architecture. Nick speaks at global industry events and blogs about MLOps, data science and ML Engineering.