Contact • Portfolio

Abstract:

Beyond the established approaches to AI alignment, newer proposals have been advanced based on needs theory, including an approach based on Maslow’s theory of motivation. While developing the ‘needs-alignment’ approach is worthwhile, in current research I am focusing on something more foundational than the Maslowian needs: the Maslowian preconditions necessary for the pursuit of needs. The preconditions illuminate important moral and political dimensions of AI by pointing to the importance of society’s normative foundations and norms of social interaction. Thus, they are potentially useful in developing a response to AI’s problematic sociopolitical implications, especially challenges to our democratic institutions.

Contact

Abstract:

Mobile sensors, such as drones, provide autonomous detection and surveillance of physical phenomena. Particularly, those which are dangerous or occur in remote areas, such as wildfires. Detecting wildfires, and estimating their parameters, in the early stages is crucial for protecting lives and property, allowing for time-critical decisions to be made regarding fire fighting and evacuation strategies. I will present some work that studies multi-drone information fusion using time-series shapelets and spatial constraint satisfaction to perform agile redeployment of drones within the mission area. Shapelets are used to extract the temporal patterns within real-time thermal data to classify the stage of combustion a drone is witnessing at a specific time and place. The temporal features are then shared amongst the drones, allowing an environmental model to be computed through spatial constraints. The resulting model is used to actively retarget and continuously redeploy the drones within the sensing field to reduce uncertainty regarding the fire’s parameters. I also hope to briefly mention a national testbed located at NCSU that can enable experimentation in support of such cyber-physical research.

Contact • Lab

Abstract:

Natural photosynthesis supports global production of food, bioenergy, and other useful products, but it has a low overall efficiency. Plants and other oxygenic phototrophs inefficiently use high-energy photons in the green, blue, and near-UV portion of the solar spectrum, and they cannot access the abundant solar energy in the near-infrared. Our lab investigates new strategies to increase photosynthetic yields to keep pace with rising global demand for food and energy. We use AI-guided de novo protein design to engineer new high-efficiency photosystems that assemble electron-transport chains comprised of bio-available redox cofactors. These de novo photosystems will be able to incorporate near-infrared light into oxygenic photosynthesis and reduce energy losses from absorption of high-energy photons. We produce synthetic genes and express the designed proteins in E. coli, photosynthetic bacteria, and plants with the goal of in vivo self-assembly and function. We characterize the protein structures and functions experimentally using high-throughput screening, spectroscopy, electrochemistry, and X-ray crystallography. If successful, our de novo photosystems will increase overall solar-to-chemical energy conversion efficiency for enhanced yields of food crops, renewable solar fuels, and sustainable ammonia for fertilizer.

Contact • Findings

Abstract:

North Carolina’s Outer Banks region offers many memorable travel experiences including crossing several miles-long bridges, taking your car on a ferry, and driving narrow, sand-lined roads that are constantly being reclaimed by the ever-shifting dunes. One of the most remarkable can be found at the northern tip of NC Highway 12, past Corolla in Currituck County, where the pavement ends and the 11-mile beach area known as “Carova” or “the 4×4” begins. Keep going onto the sand–in a four-wheel drive vehicle with aired down tires, of course–and you’ll witness pristine beaches, wild horses, protected wildlife areas, a surprising number of homes, and usually quite a few visitors. But just how much traffic does the 4×4 Beach Area see? And what impact does this unique area have on the local and state economy? Currituck County turned to researchers from NC State University’s Institute for Transportation Research and Education (ITRE) to find out.

Contact

Abstract:

How does creating with AI shape the way we imagine, express ideas, and learn? My research explores the intersection of creativity, learning, and artificial intelligence, examining how learners develop ideas and make creative decisions while working with AI. By connecting art-making with STEM learning, this work explores how creative expression can help learners engage with computational and mathematical ideas. Through the design and study of creative learning environments, my team investigates how AI can support experimentation and reflection while keeping learners’ voices and choices central. Drawing on projects involving art-making and AI chatbot design, this presentation will explore both the possibilities and tensions of human–AI collaboration: When does AI open up new creative directions? How can learners maintain ownership of their work? And how can we design learning experiences that connect artistic expression with STEM inquiry while nurturing creativity and a critical understanding of AI?

Contact • LinkedIn • Google Scholar

Abstract:

The rapid growth of IoT sensors and bioelectronic devices demand power sources that are safe near skin and in the environment, yet conventional batteries depend on flammable, toxic electrolytes that pose safety and disposal risks. This talk presents fluid-activated batteries that use water, the safest electrolyte, in two complementary forms. Water-activated batteries remain dry until a drop of water or bodily fluid hydrates the separator, giving negligible self-discharge on the shelf. At roughly one dollar per cell, they perform comparably to, and in some cases better than, commercial batteries under standard operating conditions. We have applied them to electronics-free smart bandages that deliver electrotherapy for wound healing. Building on this, moisture-activated batteries combine the functions of an energy harvester and a battery by drawing water directly from ambient air. A moisture-capture zone converts vapor to liquid, which dissolves pre-loaded salt to form the electrolyte in situ. The resulting soft, flexible device delivers about 1.6 V, powers a pulse oximeter, and enables a moisture-triggered kill switch that destroys electronics to prevent tampering. Together, these nontoxic, biocompatible platforms offer a practical power solution for wearables, medical patches, environmental sensors, and secure IoT.

Research Articles

• Safe, high-performance, moisture-activated batteries for powering next-generation Internet-of-Things devices | Science Advances
• Water-powered, electronics-free dressings that electrically stimulate wounds for rapid wound closure | Science Advances

News

• Researchers Create Moisture-Driven Tech That Powers Green Batteries – and Dissolves Spy Gear | NC State News
• This new stretchable battery harvests water from air
• Researchers create moisture-driven tech that powers green batteries — and dissolves spy gear | Rice News | News and Media Relations | Rice University
• Altmetric – Safe, high-performance, moisture-activated batteries for powering next-generation Internet-of-Things devices
• Altmetric – Water-powered, electronics-free dressings that electrically stimulate wounds for rapid wound closure

Contact • Lab

Abstract:

The ultimate mission of the newly launched Field Automation, Robotics, and Machinery (FARM) Lab is to design intelligent robotic systems that mitigate critical global challenges, such as extreme weather volatility, labor shortages, and severe resource constraints. To achieve this mission, our research focuses on developing Generalizable, Collaborative, and Scalable Robot Autonomy in challenging and unstructured field environments, with applications in, but not limited to, agriculture, forestry, and livestock. Our lab designs and deploys diverse autonomous intelligent systems across heterogeneous platforms, including wheeled ground vehicles, legged robots, aerial drones, and humanoid robots.

Contact: Jeremy Leoard (ORCID) • Sana Ahrar (ORCID)

Abstract:

Housing attainability in the Research Triangle is a wicked problem with overlapping constraints. Market pressure and regulatory friction compound rising land costs, while regional growth outpaces what gets built. The housing that does get built follows a narrow set of predetermined templates, and those templates limit who can afford to live here. This research asks what is constraining housing production in the region and which responses can work within those constraints, or push against them.

Our research is a feedback loop. In Spring 2026, we began defining the boundaries of the problem through a review of current housing policy and structured interviews with practitioners. In Fall 2026, we co-organized a Housing Symposium in the School of Architecture, bringing together architects, developers, planners, and city officials to explore the challenges and possibilities for housing production in the Triangle and beyond.

In Spring 2027, these findings will move into a graduate design studio taught in collaboration with a national architecture practice, where students will develop housing models beyond the limited set in circulation. The final phase synthesizes the constraints identified through practice with the innovations produced through teaching. Practitioners who work inside these constraints every day often see possibilities that current production models leave unexplored. By bridging practice-based challenges with graduate-level innovation, the research feeds new models back to those practicing within them.

Contact • ASCEND

Abstract:

ASCEND (Autonomous Scientific Computing Engine for Novel Discovery) is a personal, AI-powered system that automates scientific computing on high-performance clusters. It pairs a reasoning AI agent (Claude Code) with a controlled command-line harness: the agent plans and interprets, while a typed, auditable layer executes jobs on the cluster. Running on NC State’s Hazel and NCShare systems, ASCEND handles the full research loop — probing available resources, writing and validating Slurm batch scripts, submitting GPU jobs with correct hardware requests, diagnosing failures, and automatically fixing and resubmitting them. Every step records provenance, and empirical claims are verified against actual job output rather than assumed. The system runs directly on a dedicated virtual login node, keeping shared infrastructure safe while giving the agent an interactive workspace beside the researcher’s data. ASCEND has reproduced published machine-learning results end to end and captures reusable “lessons” from each run, building durable institutional knowledge. The goal is to lower the barrier to advanced computational research — letting scientists express intent in natural language while the system manages the intricate, error-prone mechanics of HPC job submission, environment setup, and reproducibility.

Contact • Lab

Abstract:

Functional magnetic resonance imaging (fMRI) is a functional neuroimaging technology that enables noninvasive mapping of human brain function. Since its advent in the 1990s, fMRI has transformed and expanded neuroscience research. However, extracting accurate individual-level information from this high-dimensional, noisy, and biologically complex data remains a major challenge. In my research and collaborations, I develop and apply statistical methods to enhance individual precision in functional brain mapping by leveraging information from the population and across the brain. These methods are designed to respect key features of the data, including its spatiotemporal structure, neuroanatomy, and individual variability in both spatial and temporal brain organization. I also develop data science tools including software, pipelines, and resource sharing, with the ultimate goal of advancing the translational impact of fMRI and functional neuroimaging more broadly.

Contact • Portfolio • Comic Experiments • Interactive Comic

Abstract:

For over a century, comic books have largely adhered to a fixed grid structure to convert space into time and navigate narrative sequence. While seminal creators like Jim Steranko challenged this formality with concepts such as the “single kinetic block”, a unified page designed to disrupt standard reading paths through random vision movement, standard digital remediation often diminishes the narrative power of the whole page through panel-by-panel viewing. This paper presents an innovative approach to digital comics that leverages the spatial integrity of the grid as the primary site of interactivity. Addressing Steranko’s unrealized intention to adapt his kinetic block for the internet, we developed an interactive project platform allowing users to physically manipulate the grid layout. We analyze specific outcomes from this experiment, most notably the work “Goodbye,” which introduces a multi-layered interactive grid where users can swap panel locations to reshape the chronology. Crucially, this innovation allows panel content to change dynamically based on its relative position to specific narrative “anchor” panels. This interaction moves beyond mere branching narratives, offering a new form of “co-authoring” where the juxtaposition of space and digital state changes deepens the emotional and intellectual engagement with the story, offering a durable mutation for digital sequential art.

Contact • Coauthor: Iseul Choi

Abstract:

Federal financial sanctions are intended to encourage state governments to address administrative deficiencies and strengthen accountability. However, states may respond differently to similar federal pressures depending on their administrative capacity and organizational practices. This study examines whether federal Medicaid financial sanctions are associated with stronger state enforcement activities and how administrative resources shape these responses. Using panel data from 50 states between 2011 and 2025, the study investigates the relationship between federal Medicaid disallowances and subsequent enforcement activities by Medicaid Fraud Control Units (MFCUs), focusing on three stages: investigations, indictments, and convictions. The analysis employs state and year fixed-effects models and Poisson regression models to examine how enforcement responses vary with staffing levels, program integrity resources, and organizational practices. Findings suggest that federal sanctions do not produce uniform improvements in state enforcement. Instead, their relationships with enforcement activities vary across stages and levels of administrative capacity. Greater MFCU staffing strengthens the relationship between federal sanctions and subsequent convictions, while greater program integrity resources strengthen the relationship with investigations. Organizational practices exhibit less consistent moderating effects. These findings highlight the importance of administrative capacity in shaping state responses to federal oversight and suggest that financial sanctions alone may be insufficient to improve enforcement effectiveness.

Contact • Findings

Abstract:

Similar to stakeholders’ demands for and the profession’s response to fraud detection in the 2000’s, investors are now increasingly interested in C-suite employees’ and entity’s illegal acts, finding them both relevant and material. This study identifies a gap between auditors’ and investors’ relevance and materiality judgments of a CEO’s illegal act. Varying role (auditor vs. investor) and using experience as a control (audit vs no audit), results show that non-auditor investors rate a CEO’s illegal act as significantly more relevant and material than do auditors placed in either role. Additionally, auditors in an investor role find a CEO’s illegal act significantly more material than auditors placed in an auditor role. Our findings highlight challenges for the audit profession in meeting investors’ expectations.

Contact • Study

Abstract:

This research investigates post-surgical breast cancer survivorship, with a focus on personalized external breast prostheses (EBPs). Integrating human-centered research with 3D scanning, parametric design, artificial intelligence, and digital fabrication, the project explores more adaptive and individualized solutions for post-surgical fit, comfort, symmetry, and everyday wear.

Contact

Abstract:

This scholarship interrogates the cultural branding of mascots in the National Football League (NFL). Adding to a wider body of knowledge that interrogates the usage of anti-Indigenous mascots and sport brands, I draw upon settler-colonial studies and the white racial frame to explore the brand narratives surrounding the racialized project of settler-colonialism in what is now the United States. Specifically, I draw from the settler-colonial narrative of Manifest Destiny to explore the framing of the New England Patriots, Dallas Cowboys, and San Francisco 49ers. Conceptualizing settler-colonial narratives enables sport management scholars, practitioners, and stakeholders to more legitimately engage colonial politics in and through professional sport organizations.

Contact • Lab • Gift Seed Project

Abstract:

Fraser fir (Abies fraseri) is the premier Christmas tree species in the US due to its natural shape, festive fragrance, and elite post-harvest needle retention. However, Fraser fir faces significant challenges due to the impacts of accelerated climate change. Several pests and pathogens already pose a significant threat to the continued success of the Fraser fir industry and the sustainable US supply of real Christmas trees. Therefore, transformative and rapid genetic improvement approaches are needed to overcome these compounding challenges and preserve Fraser fir for future generations. Domestication and genetic improvement of forest trees for human uses are primarily carried out through traditional breeding programs and can take between 20 – 30 years between generations. Alternatively, genetic engineering can rapidly improve (1-2 years) forest tree productivity and survival in the short-term. However, adoption of genetic engineering to rapidly domesticate forest trees has been limited. Development of pest and disease resistant conifers or even self-lighting bioluminescent trees through biotechnology approaches hold significant promise for future outcomes of the US Christmas tree industry. The integration of advanced biotechnology tools into traditional forest tree breeding programs will be transformative for forest tree domestication.

Explored the science of how brains process story-driven information. Practiced crafting a clear, three-sentence story about your own research in an interactive setting.

• View slides
• Hosted by:
  • College of Sciences
  • Arts NC State
  • Office of Outreach and Engagement
  • NC State Libraries
  • Office of Research and Innovation

Discovered OUR funding, grant support, and mentor training programs designed to help you recruit, develop, and celebrate student researchers.

• View slides
• Hosted by:
  • Office of Undergraduate Research

Built your knowledge of working with or sharing secondary data. Learned how to manage data permissions and ethics while navigating NC State processes and key rules like HIPAA and Export Control.

• View slides
• Hosted by:
  • IRB

Discovered how AI tools transformed agricultural research through faculty talks and a multi-perspective panel on real-world farming, industry, and policy impacts.

• Hosted by:
  • College of Agriculture and Life Sciences
  • Plant Sciences Initiatives

A practical session for researchers that addressed the use of AI throughout the life of a research project.

• View slides
• Hosted by:
  • IRB

Gained practical insights on identifying Department of Defense (DoD) funding opportunities and leveraging NC State resources for competitive proposals.

• View recording
• Hosted by:
  • Defense and Security Institute (DSI)
  • Proposal Development Unit (PDU)

Learned how High Throughput Computing (HTC) and the OSPool can accelerate your research. Ideal for workloads in image analysis, bioinformatics, AI/machine learning, model optimization, and large job sweeps.

• View recording
• View slides
• Hosted by:
  • NC State Research Facilitation Service

Explored the nuclear forensics lab.

• Hosted by:
  • College of Engineering
  • Department of Nuclear Engineering

Discovered the benefits, challenges, and practical frameworks for analyzing secondary qualitative data like archived interviews and focus groups to elevate your research.

• View recording
• View slides
• View links from slides
• Hosted by:
  • EQUAL community (Elevating Qualitative Research @ NC State)
  • NC State Libraries
• Resources:
  • ICPSR data database
  • ICPSR Summer Program
  • ICPSR’s social media data archive (SOMAR)
  • ICPSR’s Teach with Data webpage
  • Sage Research Methods
  • NC State Data Use Agreements
  • Ithaka S+R studies
• Contact: Lynnee Argabright, Research Librarian for the Social Sciences, for questions about qualitative research design, ICPSR, restricted data access, or the EQUAL community

Interacted with experts to tackle coastal NC challenges through design thinking, and learned how past attendees carried forward their innovative ideas.

• Hosted by:
  • Climate and Sustainability Academy

Gained insights from faculty experts on pressing eastern wildfire challenges—including climate impacts, human health, and fuels management—and participated in an interactive audience Q&A.

• Hosted by:
  • College of Natural Resources

Saw how the Bioinformatics Research Center connects campus computing resources through Nextflow processes, containers, and live research use cases.

• Hosted by:
  • Office of Information Technology
  • NC State Research Facilitation Service

Visited the new CPS lab space, learned about ongoing research and discussed collaboration opportunities.

• Hosted by:
  • College of Engineering

Gained interdisciplinary insights from coastal engineering, policy, and insurance experts on modernizing legacy flood risk systems for proactive and equitable disaster management.

• View recording
• Hosted by:
  • Center for Geospatial Analytics
  • Climate and Sustainability Academy

An interdisciplinary, 36-hour online “hackathon”-style weekend where high school students worldwide researched, planned, developed, and pitched a solution to a global challenge in line with the UNSDGs while working with faculty mentors and cross-collaborating with an international team.

• Hosted by:
  • Office of Global Engagement – Global Training Initiative (GTI)