Owen M. Hoffman

I am a senior at Swarthmore College completing dual degrees in Computer Science and Engineering, with research focused on human-centered safety in artificial intelligence systems. In the Collective Resilience Lab, I investigate LLM-based interventions for online safety, resulting in a first-author publication at CHI ‘26 on conversational agents for scam prevention education. My research objective is to bridge insights from human-centered HCI safety into embodied autonomous systems, particularly focusing on perception-conditioned decision making for robots that operate alongside people in unstructured environments.

Research Interests: Human–robot interaction, safe autonomy, perception-conditioned decision making, learning-based control, and human-centered evaluation methodologies for embodied AI systems.

Current Research: Research Assistant, Collective Resilience Lab (Prof. Sukrit Venkatagiri); developing and evaluating safety-focused LLM systems with applications to conversational user interfaces and online harm prevention.

Technical Background:

• First-author publication at CHI ‘26 on LLM-based conversational agents for scam prevention
• Robotics systems experience: perception pipelines, autonomous navigation, and feedback control (TurtleBot2)
• Coursework foundation in robotics, computer vision, machine learning, control systems, and digital signal processing