OPHT 116: Symbiotic Projects - Prototyping Human-Machine Systems

3 units · Letter or Credit/No Credit

Symbiotic Projects - Prototyping Human-Machine Systems is the second course in the three-quarter Symbiotics sequence from the Stanford Human Perception Lab. Students learn to build a system that senses something about a person, infers something from that signal, and adapts in response. In the first weeks every student completes the same small starter project from a provided product requirements document, using the lab's sensing tools, so that everyone has taken one system from requirements to working demonstration. Students then write their own requirements document and build their own project, individually or in a small team, toward weekly milestones. A project may be a working prototype or a research study aimed at a specific publication. Students build with frontier technologies, including new sensing hardware, current AI models, and the lab's own research tools, and learn to judge what each is ready to support in a prototype. The course draws on human-centered design methods: defining the person and the need, prototyping early, and testing with people. At each step it asks what the evidence supports: how reliable the signal is, what else could explain it, and whether the adaptation leaves the person in control. Topics include product requirements, wearable and sensor data, psychometric measures of perception and performance, signal processing, personal baselines, simple predictive models, adaptive interfaces, and privacy, consent, and human-subjects review for human data. The quarter ends with a live demonstration and a short written evaluation, leaving each project ready for refinement in OPHT 117. Prerequisite: OPHT 115 or consent of instructor. Programming experience is helpful but not required. Learning Objectives: By the end of this course, students will be able to: 1. Translate a human need into a product requirements document with milestones and measurable success criteria. 2. Collect, clean, and document human data from sensors, wearables, or software, with appropriate consent, privacy protections, and human-subjects review. 3. Build and test a model that infers a human state or intent from measured signals, and state its assumptions and error. 4. Implement a closed Perception-Decision-Action loop in which a system adapts to a person and the person can see and override the adaptation. 5. Evaluate a working prototype against the Laws of Symbiotics and report its limitations honestly.

Offered in Winter 2027 at Stanford University.

Winter 2027 sections

More OPHT courses

All OPHT courses · All departments