Diabetes is a chronic disease that requires regular monitoring of blood glucose levels to prevent serious complications. This study aims to design and test a non-invasive blood glucose monitoring system based on the Internet of Things (IoT) as a more comfortable and sustainable solution, while also supporting the achievement of Sustainable Development Goal (SDG) point 3: good health and well-being. The system uses the MAX30102 sensor, which utilizes a combination of LED and photodiode to detect glucose levels through light reflection from the fingertip, without the need for blood sampling. The measurement data is transmitted via Wi-Fi using the and displayed on the Blynk platform as well as a 16x2 LCD for real-time monitoring. This study employed an experimental method by comparing the measurements of the proposed device with a conventional device on 10 respondents in two conditions: fasting and two hours after eating. The system was tested under two scenarios—after 8 hours of fasting and 2 hours post-meal—with total of 10 participants. The average measurement accuracy of the system was 91.0% during fasting and 89.7% two hours after eating. The overall average error was 6% during fasting and 5% after eating. The system demonstrates strong potential as an early detection tool that is painless and suitable for further development.