Through the course of developing a game, AI systems often go from good ideas, to compelling implementations, to complicated frameworks, to monstrosities which are brittle and difficult to debug. This session will show you how to engineer AI systems which remain stable and robust even in the face of changing requirements, and how to use specifications and automated testing to maximize your flexibility throughout the development cycle.

Takeaway

Attendees will learn the benefits of developing stable, testable AI systems to an exact specification, and techniques for doing so. They'll learn how to limit the complexity that arises from interactions between systems, and how to cope with changing requirements without compromising stability.

Intended Audience

Not much concrete, specific AI knowledge is necessary, but this is a talk for experienced, senior AI devs looking to make their next development cycle smoother and more effective.