Back

Speaker "Rustem Feyzkhanov" Details Back

 

Topic

Serverless architecture for AI applications

Abstract

Serverless architecture for AI applications Deep and machine learning becomes more and more essential for a lot of businesses for internal and external use. One of the main issues with deployment is finding the right way to operationalize model within the company. Serverless approach for deep learning provides the cheap, simple, scalable and reliable architecture for it. Serverless architecture changes the rules of the game - instead of thinking about cluster management, scalability, and query processing, you can now focus completely on training the model. The downside within this approach is that you have to keep in mind certain limitations and how to integrate your model in a right fashion.

Who is this presentation for?
My talk will be beneficial for data scientists and for machine learning engineers.

Prerequisite knowledge:
AWS, Tensorflow - beginner

What you'll learn?
I will show how to deploy Tensorflow model for image captioning on AWS infrastructure. AWS Function-as-a-Service solution - Lambda - can achieve very significant results - 20-30k runs per one dollar (completely pay as you go model), 10k functions can be run in parallel and easily integrates with other AWS services. It will allow you to easily connect it to API, chatbot, database or stream of events. I will also show how to construct serverless workflows for deep learning which enable to conduct A/B testing of the models, Canary deployment, error handling.

Profile

Rustem Feyzkhanov is a machine learning engineer at Instrumental, where he creates analytical models for the manufacturing industry. Rustem is passionate about serverless infrastructure (and AI deployments on it) and has ported several packages to AWS Lambda from TensorFlow, Keras, and scikit-learn for ML to PhantomJS, Selenium, and WRK for web scraping.