Researching edge AI and small language models for practical deployment

The AI Exploration Center, or AIX, at SUNY Polytechnic Institute is structured as an action-driven research ecosystem focused on advancing AI innovation, promoting ethical AI development, and deepening user experience research. Its three-part model, the Laboratory, the Accelerator, and the Studio, creates a loop between theory, application, and human interaction.

As part of my graduate experience at SUNY Poly, I joined the AIX Laboratory as a Graduate Technical Researcher. My work was supported through a New York State budget allocation, which created the opportunity to contribute to open-source AI applications focused on agriculture and sustainability while participating in a broader research environment centered on practical AI development.

"The most meaningful AI research is not just about bigger models. It is about building systems that can actually operate where people need them."

I worked with Professor Dr. William Thistleton on research around edge AI and small language models that run on mobile devices. That focus came from a practical question: how do we make AI usable in contexts where cloud dependence, expensive infrastructure, and heavyweight model assumptions are barriers rather than advantages?

This work emphasized deployable intelligence over scale for its own sake. Edge AI and compact language models open up possibilities for mobile-first tools, lower-cost systems, and applications that can function more effectively in constrained environments. That direction aligned closely with my broader interest in agriculture, sustainability, and AI systems designed for underserved or infrastructure-limited settings.

In addition to my work in the AIX Laboratory, I also collaborated with the AIX Studio, where the focus shifts from model research to interaction. There, I contributed to the development of open-source UI frameworks for AI-powered web and mobile applications, drawing on my UX and UI design background to think about clarity, accessibility, and user-centered interaction patterns.

What made the experience distinct was the integration of technical and design perspectives. The lab environment created space to explore scalable and responsible AI systems, while the studio context made it possible to examine how those systems are actually experienced by people. That combination reinforced a research approach grounded in both technical feasibility and human usability.

The broader AIX model also matters. The Laboratory advances neural networks and language-model research, the Accelerator bridges theory into real-world organizational use, and the Studio studies how people engage with generative tools. Working inside that ecosystem meant the research was never isolated from practical application.

For me, this case study represents more than a research appointment. It reflects a graduate-level shift toward building AI systems that are open, applied, and deployable in realistic conditions. It also deepened my interest in the intersection of edge computing, smaller models, sustainability-focused applications, and the interfaces through which people actually use AI.

Looking back, the value of the experience was not only in the technical investigation of edge AI and mobile language models, but in contributing to a collaborative research environment where faculty, students, and partners were working toward more useful and responsible AI systems.