Building Google's ML Engine from Scratch w/ GPUs, TensorFlow, Kafka, Kubernetes
Details
This month we're in for quite a treat as we have Chris Fregly (https://www.linkedin.com/in/cfregly/), Spark contributor and founder of PipelineAI, visiting us from the bay area to give a talk on building end-to-end production machine learning models using TensorFlow, Kafka, Kubernetes, and Istio traffic routing. Chris is a great speaker that has graced Austin Meetup groups in the past, including some great Data Day Texas presentations. This is not one to miss!
The Amazon office is located up north by the Domain at 11601 Alterra Parkway (not to be confused with our 2nd building at 11501 Alterra). There is an attached parking garage with free parking. When you arrive, take the elevator to the 5th floor and check in with security.
6:00pm: Doors open, food and drinks provided
6:30pm: Talk begins
Title: Building Google's ML Engine from Scratch w/ GPUs, TensorFlow, Kafka, Kubernetes, and Istio Traffic Routing.
Abstract: Applying my Netflix experience to a real-world problem in the ML and AI world, I will demonstrate a full-featured, open-source, end-to-end TensorFlow Model Training and Deployment System using the latest advancements from Kubernetes, Istio, TensorFlow, and GPUs.
In addition to training and hyper-parameter tuning, our model deployment pipeline will include continuous canary deployments of our TensorFlow Models into a live, hybrid-cloud production environment.
This is the holy grail of data science - rapid and safe experiments of ML / AI models directly in production.
Following the success of the Netflix Culture Slide Deck (https://www.slideshare.net/reed2001/culture-1798664/2-Netflix_CultureFreedom_Responsibility) that I lived and breathed, I give Data Scientists the Freedom and Responsibility to extend their ML / AI pipelines and experiments safely into production.
Offline, batch training and validation is for the slow and weak.
Online, real-time training and validation on live production data is for the fast and strong.
Speaker Bio: Chris Fregly is the founder of PipelineAI (http://pipeline.ai/), an Apache Spark Contributor, a Netflix Open Source Committer, as well as the Organizer of the global Advanced Apache Spark Meetup. Previously, Chris was a founding engineer at IBM's Spark Technology Center, a Data Solutions Engineer at Databricks, and a Streaming Data Engineer at Netflix.
When Chris isn’t contributing to Spark and other open source projects, he’s creating book chapters, slides, and demos to share knowledge with his peers at meetups and conferences throughout the world.
