Ambient air pollution is one of the leading contributors to adverse health effects in humans. Epidemiological studies across the world have shed light on the ramifications of exposure to polluted air and even linked it to increased mortality and morbidity. Assessing the quantum of exposure to individuals would help policy makers and public in making informed decisions to combat air pollution and protect themselves. However, this would require air quality measurements at high-spatio-temporal resolution. At present, air quality is measured through sparsely located static monitoring stations. There are large spatial gaps in information captured by such a monitoring paradigm. This work proposes a consolidated framework for measuring and assessing air pollution at high spatio-temporal resolution. The first part of this work establishes an IoT (Internet of Things) based, low-cost mobile monitoring paradigm for hyperlocal air quality measurement. The second part introduces a novel multi-agent modelling framework that would allow for descriptive, predictive and prescriptive assessment of air pollution. A case study in the city of Chennai is considered to demonstrate the framework.