AI Make-up Arm Wins Hong Kong Physical AI Hackathon
SCMP · 1 SOURCESabout 20 hours ago2 MIN
Summary
Iron Bestie, an embodied AI beauty assistant featuring a robotic arm, took top honours at a Hong Kong hackathon focused on physical artificial intelligence after outperforming more than 160 competitors in a 48-hour sprint . The project stood out by combining software and hardware into a working pipeline that could listen, see, select cosmetics and physically apply lipstick to a user .
Key Points
- Organisers billed the event as Hong Kong’s first physical AI hackathon, bringing together more than 160 global competitors for an intensive 48-hour build challenge .
- Iron Bestie uses vision-language models and voice recognition to identify make-up items, match choices to a user’s calendar, and apply products directly .
- The team said local redaction middleware was built in so sensitive personal schedule and routine data could remain private during use .
- Winner Luo Yuqing studied business information systems at City University of Hong Kong and said she began experimenting with AI only about a year ago .
- Teammate Zeng Yitao, a senior algorithm developer from Shenzhen’s Southern University of Science and Technology, built the motion-control algorithms behind the robotic arm .
Why It Matters
The win highlights how embodied AI is moving beyond chatbots into machines that can perform delicate real-world tasks through integrated perception, planning and motion control . It also suggests Hong Kong’s hackathon scene is becoming a testing ground for practical robotics ideas that blend consumer lifestyle applications with privacy-aware AI design .