Jupyter Notebooks on GCP (Development Best Practices/Tooling)

Description: The Jupyter Notebook is an invaluable tool for quick prototyping and development, but lacks the best practices and processes common to software engineering. In this talk, we will showcase how to integrate these generally omitted best practices to convert your notebooks into fully reproducible, production-ready artifacts. First, we will start by creating a self-contained notebook on a local laptop. Then, by leveraging Google Cloud Platform’s Deep Learning Environment, we will showcase how this notebook, without modification, can be submitted to GCP for background execution, or transferred to AI Platform Notebooks for continued editing. Finally, we will demonstrate how to automate the creation of derivative artifacts from the notebook using a CI/CD pipeline.

Speaker Bio: Viacheslav Kovalevskyi is a Google Cloud AI Tech Lead with 7+ years of experience. Working on Deep Learning Images For Google Cloud AI. Teaching Java courses. Blogging about Deep Learning and TensorFlow. Making things work. You can find more about his work at http://tf.guru and follow him on twitter https://twitter.com/b0noi

Agenda
5:45 - 6:30 PM Networking/Social/Food
6:30 - 6:35 PM Introduction
6:35 - 7:30 PM Talk and Q&A
7:30 - 8:00 PM Networking/Social/Closing