More talks in the program:
18:15 - 19:15
Artificial intelligence (AI) and Machine Learning (ML) offer incredible opportunities for enterprises to introduce new business models, optimize their offerings and interactions with their end users, improve customer experience, and increase efficiency of their business processes and operations.
Kubernetes combined with Serverless/Function-as-a-Service (FaaS) offer the perfect stack for creating a production-ready ML framework that can power a myriad of applications and use cases within the organization — supporting granular scalability, ease of use, and portability across mixed environments spanning cloud resources as well as on-prem datacenters.
Through a live demo of a sample use case, this talk covers the suggested architecture and design patterns for enabling a distributed, scalable, ML framework that can be consumed (in a self-service/API) by various stakeholders/apps – enabling them to easily leverage ML models and data in a reliable way.
We share best practices around the various components of the stack- comprising of a managed Kubernetes solution, open source Serverless framework, data streams integrations, stateful data store recommendations, as well as key consideration for Day2 operations and maintain-ability.