Explore the intricacies of building and operating an open-source data science platform in this comprehensive workshop led by Jörg Schad, Head of Machine Learning at ArangoDB. Delve into the entire deep learning pipeline, from exploratory analysis to model deployment and monitoring. Learn how to enable data scientists to develop models exploratively, automate distributed training and serving using CI/CD, deploy frameworks on various infrastructures, manage multiple deep learning frameworks on a single cluster, store and serve models at scale, track essential metadata, and monitor pipeline performance. Gain hands-on experience constructing an end-to-end data analytics pipeline, incorporating tools such as TFX, Kubeflow, Airflow, Apache Spark, Jupyter Notebooks, TensorFlow, Jenkins, Argo, and more. Acquire valuable insights into pipeline orchestration, data preparation, distributed training, automation, model storage, serving, and monitoring throughout this intensive 2-hour and 57-minute session. Read more