How We Built Google Tulip by Using Serverless Technology and Machine Learning

Explore how Google Tulip was built using serverless technology and machine learning in this conference talk from GOTO Amsterdam 2019. Discover the process of composing an application with multiple serverless components and learn how to train an ML model with minimal data for practical application. Gain insights into the Tulip Translator project, including data gathering, model training, evaluation, and testing. Delve into the technical aspects of development using Cloudrun, Docker, Container Registry, and Build Trigger. Understand the implementation of Dialogue Flow and see a live demo of the application. Follow along as the speakers discuss the creation process, deployment strategies, and real-world applications. Access the complete talk, including slides and additional resources, to enhance your understanding of serverless technology and machine learning in innovative projects.

Introduction
Meet Matt Feigal
What is Tulip Translator
Machine Learning
Gathering Data
Training the Model
Evaluating the Model
Testing the Model
Future of Google Tulip
Cristian Hees
The most important part
Im a developer
Cloudrun
Docker
Container Registry
Build Trigger
The Real World
Deployment
How does this work
Dialogue Flow
Demo
How we built
Creating a request
Wrapping up
Github