This article investigates machine-learning concepts using resource-constrained, low-power microcontrollers; collectively termed TinyML. Machine learning continues to make its mark on many aspects of our daily lives, whether at home, in the office, or in-between. While many machine learning applications require significant computational power to crunch complex scientific or financial data, those designed for the Internet of Things (IoT) and other edge-based applications offer only meager compute and connectivity capabilities.