There is a growing demand for farm management systems that can track plant growth, allow automatic control and aid in real-time decision making to make indoor farms more viable. IoT is being applied to meet these needs and numerous researchers have created prototypes for specific purposes with sensors, algorithms, and automations. There is, however, a critical need for comprehensive large-scale experiments to test the aspects of such systems’ availability, scalability, reliability, especially the low-cost ones. The purpose of this study is to develop a low-cost IoT ecosystem using off-the-shelf components in an indoor farm setting to move from a prototype to a real-world testbed for experiments and address some of the challenges identified. Additionally, to utilize the data captured for intelligent farm management. This testbed realized collection and monitoring of nutrient quality, light intensity, environmental variables, images, and corresponding actuation of LED, irrigation, camera modules using commercially available sensors, components and Raspberry Pis, Arduino microcontrollers for computing. It featured a cloud-based dashboard for remote monitoring and control. Novel techniques of edge-computing, anomaly detection, machine vision were applied. The developed testbed is used to successfully investigate the effects of nutrition, lighting, and density on growth of microgreens guided by a statistical design and can motivate, serve as a reference to interested growers. A framework for anomaly detection and yield prediction in the farm is proposed. This study complements and expands the previous works on building low cost IoT farm systems. While the experience with the testbed indicates its importance in conducting research in a practical setting, it also points to some major lessons from the field which are described in detail.