News

Zavier N. Ndum has successfully defended his PhD thesis titled “A Generative AI–driven Framework for Robust Automation of Nuclear Modeling and Simulation Workflows with Multi-modal Domain Knowledge Integration” in March 27, 2026, as a first PhD graduate from SMART Lab.

During his time in the SMART Group, Dr. Ndum produced a remarkable body of work applying GenAI to nuclear engineering, developing novel AI systems including AutoFLUKA, RADIANT-LLM, AutoSAM, and AROMA-GPT, each of which instantiates a unified human-in-the-loop framework for automating nuclear reactor modeling, simulation, and knowledge management workflows. His research resulted in two published peer-reviewed journal articles, in Progress in Nuclear Energy (2026) and Energy and AI (2025), with additional manuscripts currently under review. Along the way, he earned several competitive honors, including the Texas A&M Chevron Energy Graduate Fellowship, the ANS Graduate Scholarship, the Health Physics Society Fellowship, the TAMIDS Data Science and GenAI Ambassadorship, the Best Oral Presentation Award at the 2025 Texas A&M Energy Research Society Conference, and the J.D. Williams Student Paper Competition Award at the 2025 INMM Annual Conference. We are incredibly proud of everything Zavier has accomplished and wish him all the best in the next chapter of his career.

Congratulations Dr. Ndum!

We are proud to announce the establishment of the Generative AI for Science and Engineering (GAISE) Lab, recently selected as a 2025 Thematic Lab by the Texas A&M Institute of Data Science (TAMIDS). Chosen through a highly competitive review process involving 15 exceptional proposals, the GAISE Lab positions our group at the forefront of AI-driven innovation.

Mission & Research Focus Led by Dr. Yang Liu, the GAISE Lab unites experts in nuclear engineering, computer science, and materials science to develop foundational AI models that integrate complex graph data, large-scale simulations, and expert domain knowledge.

Supported by TAMIDS seed funding and resources, the lab will pursue two primary pilot projects:

• Nuclear Foundation Models: Utilizing Large Language Models (LLMs) to streamline licensing, anomaly detection, and cyber-threat identification in advanced nuclear technologies.
• Materials Discovery: Developing Graph Network Foundation Models to enhance the accuracy and efficiency of predicting material properties, driving innovation in energy systems.

The Team This interdisciplinary initiative features collaboration between Dr. Liu and co-directors Dr. Shuiwang Ji (CSE), Dr. Raymundo Arróyave (MSEN), and Dr. Xiaofeng Qian (MSEN). Together, the GAISE Lab is dedicated to bolstering national energy security and training the next generation of the AI workforce.

Texas A&M Institute funded GAISE lab, which is a multidisciplinary

SMART lab welcomes new members in Fall 2025:

Akzhol Almukhametov joins the lab as a PhD student, he obtained M.S. degree from Polytechnic of Milan.

Abhiram Garimidi continues his work with the lab as a M.S. student after graduated from TAMU NUEN.

SMART Lab PhD student Zavier Ndum Ndum wins the INMM Best Student Paper Award. The work received such as recognition is “RADIANT-LLM: Retrieval-Augmented Domain Intelligent LLM Framework for Safe, Secure, and Safeguarded Design of Advanced Nuclear Reactor Technologies”. The work is collaborated with TAMU faculty members John Ford, Jian Tao, and Mansung Yim.

Kudos to Zavier!

On August 21, 2025, Dr. Liu delivered an invited graduate seminar to the Nuclear Engineering Program at the University of Florida. His talk was titled “Scientific Machine Learning to Support Advanced Nuclear Reactor Development and Deployment.”

This marks Dr. Liu’s third invited talk since establishing the SMART Lab at Texas A&M University. His other recent presentations include:

• June 2025: A talk for the United Kingdom’s Nuclear Thermal Hydraulics Collaborative Computational Project (NTH-CCP).
• November 2024: A seminar for the Department of Nuclear Engineering at the University of Tennessee, Knoxville.

SMART Lab PhD student Zavier Ndum has been selected as a Texas A&M Chevron Energy Graduate Fellow for his innovative research on applying GenAI and digital twin technology to Nuclear Engineering Applications: https://energy.tamu.edu/news_item/announcing-the-2025-2026-texas-am-chevron-energy-graduate-fellows/

Congratulations to Zavier!

In the 2025 ANS Annual Meeting, Dr. Liu served as a panelist in the “Thermal Hydraulic Education: Opportunities with AI” panel, where he discussed his experience and SMART group’s ongoing on how AI tools are being used in the classroom and in research. Dr. Liu also presented the work “Advancing Nuclear R&D with GenAI: Case Studies in Thermal Hydraulics, Simulation Workflows, and Education” which summarizes several ongoing researches on GenAI and LLM agents in SMART Group.

SMART group student Zavier Ndum presented his research work on “Development and Demonstration of an AI-Assisted Digital Twin Framework for Remote Monitoring and Control of Advanced Small-Scale Reactors”.

SMART group student Zaid Abulawi presented his research work on “Multiphysics Coupled Calculations Using OpenFOAM, Serpent, and External Heat Conduction Solver for Prismatic GCRs: A Case Study on HTTR”.

SMART group students Zaid Abulawi and Zavier Ndum started their summer internship at ANL in May. They will work with Argonne researchers in the Nuclear Science and Engineering Division to integrate GenAI and AI Agents to support advanced nuclear reactor safety analysis and modeling.

Texas A&M Hosted Inaugural AI Workshop for Science and Engineering in May. (https://engineering.tamu.edu/news/2025/06/texas-am-hosts-inaugural-ai-workshop-for-science-and-engineering.html)

Dr. Liu served as a panelist on the “AI Agents for Science and Engineering”, where he discussed the recent advancements in SMART group on using GenAI and LLM agents for various nuclear engineering applications.

Zavier, as a TAMIDS Student Ambassador, gave a presentation at the “Generative AI Mastery: From Theory to Practice“ workshop on October 12, 2024. The event, hosted by TAMIDS with collaborators from NASA, Prairie View A&M, and Texas Southern University, focused on cutting-edge AI models and their applications.
“My work on integrating Monte Carlo methods with generative AI for optimizing radiation detection systems automating Modelling and Simulation Software for advanced nuclear reactor technologies garnered valuable feedback, helping me refine my models and gain deeper insights into state-of-the-art advancements in the field”.

Zavier delivered a seminar titled “Large Language Models (LLMs) and AI Agents Applications in Nuclear Science and Engineering” at the RAISE AI Lunch Seminar on October 30, 2024. The presentation highlighted the transformative role of AI and LLMs in automating and optimizing Monte Carlo simulations and nuclear engineering workflows. Zavier’s insights showcased how AI-driven tools enhance predictive capabilities, improve decision-making, and streamline complex engineering processes, offering innovative solutions for greater efficiency and productivity in the field.