Abstract

Software is everywhere! We use it directly or indirectly in most of our daily activities. But, unfortunately, software systems continue to contain defects that often lead to major system crashes and security breaches. In fact, fixing defects is one of the most time-consuming tasks in software development. In this talk, I will discuss how artificial intelligence can be used to help system development and operations teams build more reliable and cost-effective software products. I present an overview of an artificial intelligence tool, Commit Assistant, that our team has developed in collaboration with Ubisoft. This self-learning tool automatically catches software defects as programmers write code by analyzing patterns of previous errors. I will also discuss findings from related projects with other industrial partners, namely Ericsson and CAE, with the objective of synthesizing lessons learned and highlighting future directions and challenges.