The development of APIs always begins with the old -- and meanwhile boring -- battle of API-first or code-first. However, working with APIs involves a process that includes design, development and operation. With this in mind, along with the various ways API gateways and API platforms can support these processes, the question of how to automate these processes arises very quickly.

In this session, Daniel Kocot will show how API gateways and API platforms can positively influence the development process of APIs based on the DevOps concept and CAMS model to deliver better and better APIs to consumers.

This talk introduces Site Reliability Engineering (SRE) at Google, explaining its purpose and describing the challenges it addresses. SRE teams manage Google's many services and web sites from our offices in Pittsburgh, New York, London, Sydney, Zurich, Los Angeles, Dublin and Mountain View. They draw upon the Linux based computing resources that are distributed in data centers around the world.

As companies grow in size, achieving true agility becomes more and more challenging. Adding OKRs on top of that can create even more friction. But can OKRs really help your organisation be agile? In large technology organisations, friction is created when trying to implement and track OKRs. This happens because OKRs are perceived as a waterfall-like way of presenting business value to the teams and traditionally companies have been using activity-based OKRs. We will explore how agile can scale, by adopting value-based OKRs, while still leaving space to the teams to self-organize and plan their work. Examples will be given from both Scrum and large scale agile like SAFe.

Developers are drowning in streams of data from their products, assets, cloud services, apps and infrastructure. Most is only ephemerally useful, but data streams never stop. How can they derive continuously useful situational awareness when they are still stuck with batch-centric “store-then-analyze” architectures? Is it possible to analyze, learn and predict from streaming data on-the-fly? Fortunately, the answer is often ‘yes!’ Due to innovations in continuous intelligence and in-memory computing, it is possible for developers to make powerful discoveries directly from streaming data with applications using this approach: -Always have an answer from the latest data: Algorithms are adapted to analyze, train and predict continuously - without supervision, and analysis is driven by data. -Continuously analyze and predict: Each new event is statefully analyzed in memory, in real-time; Insights and predictions form a real-time stream. -Analyze in context: Fluid relationships between data sources - like containment, proximity, or even correlation - are critical for deep insightsDevelopers can create simple object-oriented applications in Java, Javascript or Python. In this talk, Simon will show how a simple, yet powerful applications can track satellites in orbit, predict their trajectories and show them on a 3D map. SwimOS (Swim Open Source) provides a rich suite of algorithms that developers can use to add your own analysis, including models of mathematical and geometric structures, rings, fields, vector modules and spaces, affine spaces, tensor spaces, probability distributions, and associated operators.

The hype around AI is real, but who among us is actually capitalizing on it? It's estimated that anywhere from 50-90% of AI models developed never make it past the AI "valley of death" that exists between the lab and deployment into production. This talk will cover how an API-centric approach to building and maintaining AI-enabled applications can bridge the divide between data scientists, software developers, and infrastructure managers and make the power of AI accessible to everyone. Not only is this approach better by making AI more accessible and easier to use, but also offers leaders the sophisticated governance and compliance monitoring needed for AI at scale.