Episode 2: Deploying Deep Learning models with TimTem

This Episode is recorded at the Intelligent Cloud Conference in Copenhagen. It features an interview with Henk Boelman Microsoft AI MVP and Cloud AI Architect for Heroes in the Netherlands.  We discuss a lot of really interesting subjects. Kicking off with Henk's session at Intelligent Cloud - Where he is using DevOps to accelerate models in to production. Then the amazing initiatives that Henk leads including the Global AI Bootcamp and the Global AI Night. We finish by discussing the charity that Henk and his wife set up which is looking to use AI to better identify and help children who have been injured Uganda.   

In this episode we explore what is a data lake, and how to build a lake which enables data science teams to deliver models faster. We need somewhere we can store and access data which is indexed, searchable and always available. A good data lake will do this and more. Listen to find out how you can build a data lake to accelerate your teams machine learning. 

I caught up with the creator of Apache Spark and Databricks founder Matei Zaharia at this year's Big Data London. We discussed the release of MLFlow, Databricks and project Dawn. 

In this episode I talk about why you should have all your data science projects in version control / source control. I discuss why it is important, how to get started, the gotchas and how to version control data science projects. 

In this episode, I am joined by #TimTem AKA Dr Tim Scarfe and Dr Tempest van Schaik. Tempest and Tim are Machine Learning Engineers for Microsoft. This podcasts focuses on a project they delivered for Confused.com.  We discuss deploying Deep learning models in to production, NeurlIPS conference, model management, Devops and much more. 

This episode of Data Science in Production, focuses on the problems in Data Science and the rise of the Machine Learning Engineer.