QGIS is great, and for people entering the GIS rabbit hole, another rewarding acquaintance is ESA’s SNAP [1]. It is a powerful toolkit for downloading and processing satellite imagery from ESA satellites.
QGIS and SNAP fits together through their respective python bindings.
Behind both is GDAL [2] the omniconverter of coordinate systems and files.
I recently discovered Qgis as I had to put on maps various data. I was quite amazed by its capabilities while being relatively easy to learn. Definitely a great open source achievement!
It is one of the best pieces of software that I've ever used. Unfortunately, it commonly gets overlooked because of the entrenched 800lb gorilla of ESRI locking in users while they're still in college.
QGIS forms the front-end for a lot of our automation and processes. We use it as an input, usually by drawing areas of interest, and then triggers in Postgres fire off containers to process those areas and fill in contents of other layers. We also use it extensively in planning, surveying, and documentation. It features heavily in all stages of our product lifecycle, from viability planning through to in-life servicing. If not for QGIS we'd likely be paying someone like ArcGIS a lot of money for a product significantly less flexible and hackable. For interest this is for a fibre ISP.
Arcmap is flexible in that it comes with arcpy. How flexible depends on how much you pay. Unfortunately, arcpy is unreliable for some use cases. My colleagues and I found it to be the only python module ever that could result in different results for the same inputs each time it ran.
We've since switched to open source GIS and we'll never be going back
QGIS and SNAP fits together through their respective python bindings.
Behind both is GDAL [2] the omniconverter of coordinate systems and files.
[1] https://step.esa.int/main/toolboxes/snap/
[2] https://gdal.org/