Abstract Urban climate model evaluation often remains limited by a lack of trusted urban weather observations. The increasing density of personal weather sensors (PWSs) make them a potential rich source of data for urban climate studies that address the lack of representative urban weather observations. In our study, we demonstrate that carefully quality-checked PWS data not only improve urban climate models’ evaluation but can also serve for bias correcting their output prior to any urban climate impact studies. After simulating near-surface air temperatures over London and southeast England during the hot summer of 2018 with the Weather Research and Forecasting (WRF) Model and its building Effect parameterization with the building energy model (BEP–BEM) activated, we evaluated the modeled temperatures against 402 urban PWSs and showcased a heterogeneous spatial distribution of the model’s cool bias that was not captured using official weather stations only. This finding indicated a need for spatially explicit urban bias corrections of air temperatures, which we performed using an innovative method using machine learning to predict the models’ biases in each urban grid cell. This bias-correction technique is the first to consider that modeled urban temperatures follow a nonlinear spatially heterogeneous bias that is decorrelated from urban fraction. Our results showed that the bias correction was beneficial to bias correct daily minimum, daily mean, and daily maximum temperatures in the cities. We recommend that urban climate modelers further investigate the use of quality-checked PWSs for model evaluation and derive a framework for bias correction of urban climate simulations that can serve urban climate impact studies. Significance Statement Urban climate simulations are subject to spatially heterogeneous biases in urban air temperatures. Common validation methods using official weather stations do not suffice for detecting these biases. Using a dense set of personal weather sensors in London, we detect these biases before proposing an innovative way to correct them with machine learning techniques. We argue that any urban climate impact study should use such a technique if possible and that urban climate scientists should continue investigating paths to improve our methods.