Introduction to a new space: MLOps

Categories: ML AI Machine Learning MLOps Artificial Intelligence Ops Operations

Join us for a session on MLOps, where we delve into the transformative practices and tools that bridge the gap between machine learning development and production deployment. Discover how MLOps enhances collaboration, reproducibility, and scalability in machine learning projects, ensuring seamless transitions from data engineering to model monitoring. Learn about the latest technologies, including Docker, Kubernetes, and MLflow, and explore real-world case studies highlighting best practices and common challenges. Whether you’re a data scientist, engineer, or manager, this session will equip you with the knowledge to streamline your ML workflows and drive impactful business outcomes.

This presentation will assume that the attendees have little to no knowledge of creating and operationalizing ML Models.

In this presentation, we will perform a rigorous list of what is required to be successful in the MLOps space.

  • Model Development

  • Model Packaging

  • Model Deployment

  • Model Cataloging

  • Model Monitoring

  • Model Maintainance

Then, we will discuss the technologies that we can use to piece these technologies together:

Some of the technologies we will discover include:

  • Airflow, Kubeflow, MLFlow

  • Prometheus & Grafana

  • TensorFlow, XGBoost, Dask, and More

  • Serving Models

  • Kafka

  • Hyperparameter Tuning