I'm Justin Del Vecchio, teaching faculty at UB's Computer Science and Engineering. Before academia, I spent 20 years in industry as a research scientist and software engineer, working with exceptional teams on applied research projects in support of the U.S. Department of Defense. You can find my full professional background on LinkedIn.
I have a long-standing curiosity about how artificial intelligence actually gets used—especially the gap between cutting-edge research and real-world deployment. In this space, I share thoughts on AI, data-intensive systems, research projects, and lessons from the classroom, with a particular focus on how emerging tools reshape the work of engineers, scientists, and institutions.
One should use LLMs and chat agents warily. They should not be the first authority consulted. Instead, develop a decent mental model first, then consult the chat agent to refine or expand points. The paradox is that asking good questions requires knowing enough about the topic to recognize what is missing.
Read moreThis post focuses upon Title II accessibility. It documents a real world problem and proposes novel and computationally repeatable applications of AI to solve them. The application exists as an open source project that employs LangGraph and OpenRouter to convert LaTeX beamer documents to Typst, inspired by UB CATT and the need for accessible course content.
Read moreLarge language models are powerful—but the next wave of AI impact won't come from ever-larger foundation models. In this post, I explore how the Model Context Protocol (MCP) enables LLMs to work alongside small, domain-specific predictive models, unlocking valuable insights from everyday data sources like CSV files. Drawing on student projects from my Fall 2025 course, I argue that the future of AI is modular, composable, and deeply tied to the data organizations already have.
Read moreJustin M. Del Vecchio, PhD
Assistant Professor of Teaching
Computer Science and Engineering
School of Engineering and Applied Sciences