The research paper introduces the development of a soil moisture sensor in accordance with the field condition and has a capacity to measure soil moisture up to 50 cm depth. In this study, the Arduino-based soil moisture sensors such as Stainless Steel 316L grade (SS316L) probe and Copper Based Probe (CBP) are introduced, which fit best with the conventional method for measuring soil moisture. The paper also presents the calibration and performance comparison of three sensors, including SS316L and CBP sensors (developed) and Manual Soil Moisture (MSM) sensors. The globally accepted and proven Oven Dry Method (ODM) is adopted for calibration purpose during soil moisture measurement. Linear, polynomial, and logarithmic regression equations, along with Adaptive Neuro-Fuzzy Inference System (ANFIS), are utilized for the best fitting. Thus, it proves that ANFIS non-linear system gives the best result. It is observed that the SS316L sensor performs well with the desired accuracy in the given field condition compared to other soil moisture sensors. The outcome of this study will assist the government authorities, policymakers, and researchers working on water-saving objectives.