The CO2 sensor calibration thing is inherent in the technology. They drift, a lot, and without occasional reference to a known standard, there's no way to know whether "1000" is really 1000, or 500, or 2000, but exactly how that gets implemented seems to vary a lot. I have an SCD30 board from Adafruit, which internally records CO2 minima and, over the course of week or so, adjusts its calibration so that minimum is 420. That means no special calibration procedure, but it does have to be somewhere that it gets periodic fresh air exposure.

There's a newer, photoacoustic sensor technology that doesn't seem to require continuous recalibration, but (at least this one: https://www.sparkfun.com/products/22956 ) require an extensive initial calibration.