This coding challenge is a simplified real world problem we were facing during one of our projects.
This challenge is about satellite image metadata for fields. Each field is monitored by satellites which take pictures of it multiple times a week in different spectrums resulting in multiple images per shot. These images are differentiated by their layer_name (NDVI, NDWI and TRUECOLOR). The satellite also detects if there are clouds above a field and calculates a cloud_coverage percentage.
As a user I want to be able to query satellite image metadata to see what's going on on my fields. For each field I want to determine the best (in terms of cloud coverage) satellite image per layer with a weekly resolution to have a good visualization of my fields.