Cheap Code, Costly Judgment: Using Software Agents for High-Quality Software Engineering

Abstract: Generative artificial intelligence (AI) is changing software engineering from a practice constrained by scarce implementation effort to one shaped by abundant, low-cost code production. This talk examines the engineering challenge that follows: how to organize architectures, tools, evidence and feedback loops so that AI-mediated development remains auditable, correctable and maintainable. Drawing on a 16-week case study in which we used frontier coding agents to build a production-grade greenfield software system, we present an AI-first software engineering methodology for turning recurring failures into durable engineering controls.

What’s in it for you: A practical methodology for using coding agents to develop software systems whose quality, reliability and maintainability matter.

Bio: James C. Davis is an assistant professor of electrical and computer engineering at Purdue University. He worked for IBM from 2012–2015 and received his Ph.D. from Virginia Tech in 2020. His work appears at venues such as the International Conference on Software Engineering and the Institute of Electrical and Electronics Engineers Symposium on Security and Privacy, and it has been recognized with three Association for Computing Machinery Special Interest Group on Software Engineering distinguished paper awards. His lab is supported by the National Science Foundation (NSF), Google, Rolls Royce, Cisco, Socket and OpenAI. He received the NSF CAREER award in 2026.

Series: See upcoming and previous presentations at CS Seminar Series.