Program co-directors Phebe Vayanos, Andrew and Erna Viterbi Early Career Chair in Engineering, and Bistra Dilkina, Dr. Allen and Charlotte Ginsburg Early Career Chair in Computer Science.
USC Viterbi School of Engineering’s pioneering graduate program in Operations Research and Artificial Intelligence (ORAI), including a PhD certificate, has launched its inaugural semester, training graduate students at the forefront of AI for decision-making. Supported by a $2.9 million National Research Traineeship (NRT) grant from the National Science Foundation, the program officially began in August, representing the first dedicated training program designed to bridge AI and operations research (OR) through purpose-built coursework, seminars, workshops, symposia, and research experiences.
“This program is very unique. It’s the first program that’s training students as interdisciplinary researchers in this convergent area,” said Phebe Vayanos, program co-director and Andrew and Erna Viterbi Early Career Chair in the Daniel J. Epstein Department of Industrial and Systems Engineering. “This means that ORAI students will be able to solve problems other students cannot.”
Vayanos noted that the program has already generated an early success story: through the ORAI Seminar’s entrepreneurship panel, one ORAI trainee was awarded a Tech Innovation Fellowship from the Viterbi School Office of Technology Innovation and Entrepreneurship to help launch a cybersecurity-focused startup aligned with the ORAI program’s themes.
Purpose-Built Courses for a Convergent Field
A defining feature of the ORAI program is its commitment to true interdisciplinary training. Rather than combining existing courses from different departments, the USC Viterbi faculty behind the program, led by Vayanos and co-director Dr. Allen and Charlotte Ginsburg Early Career Chair in Computer Science Bistra Dilkina, developed entirely new courses specifically for this convergent area.
“Our courses are co-developed and co-taught by one faculty member in computer science and one faculty member in industrial and systems engineering—one with expertise in AI and one with expertise in operations research,” Vayanos said. “This structure is very unique and ensures students learn to integrate both viewpoints from the ground up.”
“We’re training students to become the architects of the next generation of AI—end-to-end systems where prediction and optimization are developed together, not in isolation,” Vayanos said. “These are AI systems that don’t just predict but make responsible, holistic decisions in the real world and at scale. This is the future of AI, and our students are at the front of it.”
Connecting Research to Real-World Impact
The program’s seminar series brings leading researchers to campus to highlight cutting edge research and real-world applications. For example, this Fall, Ishai Menache from Microsoft Research presented how Microsoft uses large language models to optimize its cloud supply chains and to explain the solutions returned by complex optimization models in natural language—perfect examples of the complex, data-driven decision systems that ORAI students are being trained to build. Other speakers included Professor Milind Tambe from Harvard University, Dr Serdar Kadioglu from Fidelity Investments, Professor Carla Gomes from Cornell, and Professor Van Hentenryck from Georgia Tech. All the seminars are available on the program’s YouTube channel.
A central focus of the program is designing responsible AI systems. Through optimization techniques, students learn how to address critical challenges including fairness, transparency, and robustness in high-stakes applications.
“Optimization gives us the power to hard-code fairness into AI systems—to prevent historical biases from creeping in and to stop new ones from emerging,” Vayanos said. “We’re teaching students to build models that don’t just learn from historical data, but actively resist the inequalities baked into it.”
Privacy protection is another key concern. “Machine learning models can leak private information—even if the underlying data never leaves the vault,” Vayanos said. “AI is incredibly powerful, and we want to harness that power, but never at the expense of privacy. Our students learn how to build systems that are both high-performing and protective of the people they serve.”
Preparing Students for High-Impact Careers
The ORAI program prepares students for diverse career pathways in academia and industry. Graduates will be well-equipped for positions at major technology companies such as Amazon, Meta, Microsoft, and Uber, where they can work on supply chain optimization, pricing, logistics, and the development of decision-support tools. They may also work in national research labs, non-profits, or in the public sector.
“The skills our students gain are deeply transferable—they can be applied to supply chains, transportation networks, pricing systems, public policy, and countless other domains where data drives decisions,” Vayanos said. “We teach them to fuse machine learning with optimization so they can build systems that don’t just analyze data, but act on it intelligently. And through workshops on team building, ethics in AI and OR, human-subject experiments, and entrepreneurship, they learn how to design solutions that are not only technically strong but grounded in real human needs and real human impact.”
The ORAI program also recently assembled an advisory board of leading international experts spanning operations research and artificial intelligence. Members include Daniel Kuhn of the École Polytechnique Fédérale de Lausanne, a world leader in stochastic and distributionally robust optimization and its applications in Machine Learning; Radhika Kulkarni, former INFORMS President and pioneering industry executive who helped build SAS’s Advanced Analytics division; Ramayya Krishnan, former Dean of Carnegie Mellon’s Heinz College, expert in data-driven public policy and socio-technical systems, serial academic entrepreneur, and former chair of the AI Futures committee of the National AI Advisory Committee to the President and the White House Office of AI initiatives; Tuomas Sandholm, Carnegie Mellon professor and founder of multiple AI companies whose breakthroughs in game theory and large-scale optimization produced world-champion poker AIs; David Shmoys of Cornell, a leading authority in approximation algorithms and data-driven optimization for social systems; Carla Gomes of Cornell, renowned for foundational work in AI and computational sustainability; and Kevin Leyton-Brown of the University of British Columbia, a prominent researcher in algorithmic game theory and applied mechanism design, known for major contributions to combinatorial auctions, computational economics, and AI for social impact.
Looking Ahead
For Vayanos, the most exciting aspect is the program’s pioneering nature. “It’s completely state-of-the-art—really the first of its kind and we get to shape it. I am certain that students will come to USC for this program, to get this training. And at the same time, I believe it will act as a magnet for more top faculty in this convergent area.”
She added, “We see USC becoming a global hub for research and training at the intersection of operations research and AI—a place where new ideas, new methods, and new leaders are shaped. This is just the beginning of what we can build here.”
To stay up to date with the ORAI program, follow it on LinkedIn, Instagram, YouTube, or X/Twitter
Published on December 2nd, 2025
Last updated on December 4th, 2025
