It can be difficult to model urban heat because, while temperature is regional, the actual human experience of heat is strongly affected by local urban features like shade, trees, and water. These complications in scale mean that analyses are often either generalized across a neighborhood- or city-scale, or they are very detailed but hard to repeat in multiple cities without expensive input data. Here, we create a completely open-source approach to modeling human-scale (1 m) heat exposure alongside neighborhood-scale air temperature. We first combine multiple global and regional open-access datasets to produce a three-dimensional view of the city, including building heights, land use and tree cover. We also incorporate the effects of vegetation and surfaces on air temperature. Ultimately, we produce analyses of extreme heat in cities that are detailed enough to support local policy and planning but are replicable in cities around the world.