Community Partner Summit
Wednesday, February 9, 2022
In our days, the demand for developers is increasing with more and more companies looking to hire new developers to join their projects. Sometimes this "joining" is painful, and leads to huge costs for companies, as adapting and understanding a project is not an easy thing.
Are companies aware of that cost? What can a developer do for that? Are there any best practices?
We are going to explore some best practices from 2 main aspects: a) Project Management i.e. making a Software more comprehensible and easy to understand during the whole development process and b) From-zero-to-hero i.e. exploring a whole new project relying only on any existing documentation for example on open source projects, or any legacy project (either for maintenance or for incubation)
The metaverse is many things to many people, however, always involving an interface between people and things in the real world and their simulated counterparts. As experiences move beyond gaming and entertainment, the need to have trust and firm guarantees requires us to reexamine fundamental building blocks of what powers a metaverse experience. At Ably, we focus on connecting people in realtime, allowing them to chat, interact and stream data to an unlimited number of users. What we’ve found is that traditional assumptions of availability and scale don’t hold true when connecting users to the latest trends on social media, or when a whole smart city connects to its digital twin. In this talk I will walk you through how we build for sudden scale events while maintaining a level of service that can be used to drive business decisions.
In this presentation and demonstration, attendees will learn about:
- Kubeflow 1.5 features and use cases
- How Kubeflow streamlines ML workflows and simplifies operations
- Why market leaders are building their ML Platform on Kubeflow
- Kubeflow Community User Survey results and benchmarking your ML Platform KPIs
- How to get involved in the Kubeflow Community
The demonstration will provide a brief review of valuable ML workflows i.e. the automated process to build a Kubeflow pipeline (directly from a Jupyter notebook) that will train and tune an ML model. It will also show how to deploy that model in an inference server and monitor it.
The past few years have seen the appearance of different software companies promising to augment the developer through Artificial Intelligence. Within these, many have specialised on testing problems, and in particular on test generation.
In this presentation we will tell the story of AI-generated testing, and the technologies behind it. We will look at today’s main players in the industry - within different categories of testing, talk about the limitations of this technology and of its practical use cases, and explore the opportunities for the next few years.