This study presents the development and validation of a low-power, non-invasive wrist-based device for continuous glucose monitoring, offering a more convenient alternative to traditional fingertip-based methods. By employing photoplethysmography (PPG) and the Beer-Lambert law, the prototype measures the absorbance of infrared and red light through the wrist to estimate glucose levels. A comparative analysis was conducted against a commercial glucose meter (Accu-Chek) using data from a young population in Cartagena, Colombia. The results demonstrated a moderate correlation (Pearson correlation coefficient: 0.49) between the two devices, with most measurements falling within clinically acceptable zones (A and B) of the Parkes error grid. Signal processing techniques, such as noise reduction and sampling rate optimization, were implemented to enhance measurement reliability. The study highlights challenges related to sensor placement and skin tone variability, which can influence device accuracy. While the findings support the feasibility of this wrist-based device for continuous monitoring, further improvements, including miniaturization and evaluation across diverse populations, are essential for widespread adoption. This research underscores the potential of non-invasive technologies to address the discomfort and accessibility limitations of invasive glucose monitoring methods.