Machine Learning Operations (ML Ops) is an important discipline that bridges the gap between the development of Machine Learning (ML) models and their deployment in real-world applications. We will have an in-depth look at the different components of a fully functioning ML pipeline, from data ingestion, preprocessing to model training, validation, deployment, and monitoring. We will explore how we can ensure consistent performance while re-training the model with new data on a regular basis.
The session will also highlight the role of hyperscalers such as Amazon AWS and Microsoft Azure and how we can use their services to do ML Ops in the cloud. Finally, methods to expose ML models as APIs are shown. This enables us to seamlessly integrate with the front end to ensure the ML model is used properly and can scale when demand is high.
Program:
18:00 - 18:05 Opening & Introduction (Johannes Schneider)
18:05 – 18:45 "From Concept to Deployment: The Journey of ML Ops in Modern Machine Learning. A practical View" (Jonas Bokstaller)
18:45 – 19:00 Q&A
19:00 - 20:00 Apero
