Data Science
[PAST EVENT] Schmidt Family Distinguished Talk: Douglas Schmidt
Navigating the Future of AI-Augmented Research and Education
Abstract:
My presentation examines the evolution and impact of AI-augmented research and education, particularly in the context of software engineering and AI engineering. I focus on two pivotal aspects related to research: advances needed in development paradigms and architectural paradigms. Recent studies predicted that within a span of a decade, there would be a transformative shift in software development, where collaborative endeavors between humans and AI would prioritize system building based on intent versus traditional manual coding. However, the recent advent of generative AI models is rapidly shifting this timeline to the left.
Generative AI models provide both opportunities and challenges to software and AI engineering. I concentrate on two integral domains: AI-augmented software development and the engineering of AI-enabled systems. Throughout my talk, I explore the potential of generative AI tools in aiding various software engineering stages, ranging from requirements gathering to architecture and design. A particularly promising area is 'prompt patterns" for prompt engineering, which are provide valuable knowledge transfer methods for interacting with generative AI models. I also discuss the distinct challenges posed by AI-enabled software systems, include the need for AI traceability and addressing emergent behaviors in mission- and safety-critical systems.
Finally, drawing from my personal experiences teaching upper-division computer science courses at Vanderbilt, I demonstrate the pros and cons of applying generative AI tools in the classroom. Although there are many benefits, these advances are not without challenges, necessitating a reimagining of the assessment we apply in computer science course in an increasingly AI-centric world.
Bio:
Dr. Douglas C. Schmidt holds the Cornelius Vanderbilt Professorship of Computer Science and serves as the Associate Chair of Computer Science at Vanderbilt University. He is also a Visiting Scientist at the Software Engineering Institute at Carnegie Mellon University, where he was the Chief Technology Officer from 2010 to 2012. He's also been the Associate Provost of Research and the Co-Director of the Data Science Institute at Vanderbilt. Dr. Schmidt has an extensive background in commercial technology, having served as CTO for Zircon Computing and PrismTechnologies, and in government, where he was a Deputy Office Director and Program Manager at DARPA from 2000 to 2003.
Dr. Schmidt's research focuses on software engineering, specifically in patterns, optimization techniques, and tools for developing mission-critical middleware for distributed real-time embedded systems and intelligent mobile cloud computing applications. He has published 650+ technical papers and books that have garnered 46,500+ citations and has an h-index of 91. Dr. Schmidt's research has significantly impacted the software industry through his development of the widely-used ACE and TAO open-source middleware frameworks. He holds
B.A. and M.A. degrees in Sociology from William & Mary, and M.S. and Ph.D. degrees in Computer Science from the University of California, Irvine.
This talk is open to the public and a part of the distinguished speaker series sponsored by the Schmidt Family.