University Scholar Subramanian Sankaranarayanan
The University Scholars Program, sponsored by the Office of the President, honors faculty members for superior research and teaching, along with great promise for future achievements. The award provides $15,000 a year for three years to enhance their scholarly activities.
Subramanian Sankaranarayanan
Professor, Mechanical and Industrial Engineering
UIC College of Engineering
Years at UIC: 6
What themes or questions drive your research?
I develop physics-informed artificial intelligence/machine learning approaches to accelerate materials discovery, design and reliability. My group bridges the gap between electronic, atomistic and mesoscale models with multimodal experiments to understand how defects, interfaces and metastable phases govern function — especially for microelectronics, energy storage, thermal management and catalysis.
What sparked your interest in these research areas?
Two things: a fascination to understand how simple atomistic interactions dictate complex behavior at macroscale, and the opportunity to couple high-performance computing with emerging AI techniques to turn rich experimental and synthetic materials datasets into predictive, design-ready models.
What courses do you teach, and are there topics you particularly enjoy teaching?
I usually teach Introduction to Thermodynamics and Machine Learning and Data Science for Mechanical Engineers. For the former, I focus on fundamentals as this subject is cross-cutting across nearly every engineering branch (from engines and HVAC to batteries and manufacturing). In the machine learning/data course, we design projects based on contemporary mechanical engineering and materials problems so students can understand the various facets of end-to-end machine learning pipelines — from data engineering and model selection to uncertainty and deployment — that are closely aligned with current industry demand.
What strategies help you balance teaching and research?
I aim to tightly integrate them. Class projects use open datasets or problems from our lab or current literature. Students contribute to publishable results and learn to develop design workflows using artificial intelligence and machine learning. In the classroom, we emphasize problem formulation first — clarifying objectives, constraints and assumptions — then selecting the right tools (analytical models, simulations, experiments or machine learning) and iterating with data to reach physically viable solutions. Milestone-driven mentoring, shared code and data practices and concise feedback loops keep both teaching and research moving in sync.
What advice would you give to students interested in research careers?
Build strong fundamentals and communicate clearly in both oral and written form. Collaborate across disciplines and facilities. Be agile in learning and adapting. Most important of all, be patient and persistent: Important results often come after many careful iterations and checks.
