Data from the Indonesian Nutritional Status Survey indicates that 21.6% of toddlers in the country suffer from stunting (SSGI). To help prevent stunting, anthropometric measurements are often conducted in health institutions to monitor child growth. This research aims to develop an anthropometric assessment technique equipped with a Kalman filter to reduce fluctuations and improve the accuracy of weight readings by load cell sensors. The tool incorporates an Internet of Things (IoT) based system that will automatically assess the nutritional condition of toddlers. This research contributes by integrating infant and toddler weight measurements into one compact tool, enhancing weight value accuracy, calculating z-score values automatically, and displaying a graph of weight growth compared to age and gender of infants and toddlers. The results anthropometric measurements will be displayed on Nextion device and forwarded to website, where visitors can download the data in Microsoft Excel format. Results indicated that error rate for toddler scales (standing) was 0.10%, which is lower than the average error rate for infant scales (lying) of 1.06%. The error value decreased from 0.122% before using the Kalman filter to 0.088%. The comparison between manual z-score calculation and system's calculation showed the system's z-score calculation was accurate, matching the manual calculation and correctly classifying the nutritional status of toddlers. This study has implication for assisttings health workers in monitoring toddler conditions and reducing errors in anthropometric data recording, potentially helping to reduce stunting cases. Additionally, the tool simplifies the weighing process of infants and toddlers in one compact device.