How AI and Hardware can Transform Point of Care Workflows

Explore how specialized AI hardware can revolutionize medical point-of-care workflows in this 21-minute DEF CON 33 conference talk. Learn about PamirAI's innovative layered technology approach that combines a customized CM5 board, RP2040 co-processor, and sunlight-readable E-ink display to run large language models entirely on-device alongside transcription and text-to-speech models. Discover how their "MCP Hub" software layer transforms plain-language requests like "track heart rate every five minutes" into reliable data logs that function even without Wi-Fi connectivity, and see how AI-assisted coding enables any sensor to become operational within five minutes. Gain insights from veteran engineers Kevin and Tianqi, who bring experience from Microsoft Surface devices and Qualcomm's efficient-AI initiatives, as they demonstrate how they've miniaturized enterprise-grade inference into badge-sized hardware that fits 3-billion-parameter language models into a 10-Watt, pocket-safe form factor, providing clinicians with instant, private access to AI reasoning directly at the bedside while handling the massive data streams from vital signs, imaging, and clinical notes that characterize the bio-medical industry.