Research mentorship
Advising student research in AI and robotics
Research starts to become real when an interesting idea meets a constraint. I mentor students as they narrow questions, choose technical approaches, build prototypes, and account for what their results do—and do not—show.
Students mentored annually in research
Software, perception, and physical systems
THE CHALLENGE
Move from an ambitious topic to a manageable question.
Students are often drawn to a powerful technology before they have a question they can investigate. The work of mentorship is to help them define a scope, identify a meaningful comparison, and decide what evidence would support a conclusion.
THE APPROACH
Make the technical choices part of the inquiry.
Through the Applied Sciences and Engineering Program, I advise more than 10 students annually in AI, robotics, engineering, and algorithmic research. I support problem formulation, prototyping, debugging, and interpretation while keeping the student’s contribution visible.
DIRECTIONS IN DEVELOPMENT
Perception, navigation, and human–machine interaction.
Current advising directions include visual-language-model navigation with depth information and a low-cost 3D gaze-estimation interface for a robotic arm. These are developing investigations, not claims of completed systems or validated performance.
EVIDENCE
Technical mentorship across courses and research.
At St. Paul’s, the laboratory connects computer vision, machine learning, embedded computing, and robotics. At Taft, I mentored more than 15 projects involving algorithms, mathematical modeling, and machine learning.
DESIGN PRINCIPLE
A working prototype opens the next question.
A system can work in a demonstration and still fail under different conditions. I ask students to examine those conditions, explain their choices, and make the limits of their conclusions explicit.