A software × hardware IoT system that turns invisible thermal microclimates into visible information, letting library users find their most comfortable seat — built with ESP32 sensor mesh, dual-screen React frontend, AWS-hosted Python backend.
UQ DECO7285 capstone — a software × hardware IoT system addressing the invisible thermal microclimate problem in open library study spaces. ESP32 sensor mesh + dual-screen React frontend + Python/AWS backend. I led a 6-person team as Team Lead + PM + product design lead + frontend lead; the project earned a High Distinction (7/7, max grade). Key product call: after Contextual Inquiry revealed the library couldn't actually change room temperature, I redefined the button from a 'control tool' into a 'perception recorder' — flipping the entire product narrative.
I'm writing the full English case study. The Chinese version below is the canonical write-up — covering user research, three prototype iterations, two rounds of usability testing, the IoT/React/AWS stack, and reflections on cross-domain PM work.