https://numpy.org/neps/nep-0029-deprecation_policy.html

This is a recommendation that all projects across the Scientific Python ecosystem adopt a common “time window-based” policy for support of Python and NumPy versions.

This means, for example:

On next release, drop support for Python 3.5 (initially released on Sep 13, 2015)
On Jan 07, 2020 drop support for Numpy 1.14 (initially released on Jan 06, 2018)
On Jun 23, 2020 drop support for Python 3.6 (initially released on Dec 23, 2016)
On Jul 23, 2020 drop support for Numpy 1.15 (initially released on Jul 23, 2018)
On Jan 13, 2021 drop support for Numpy 1.16 (initially released on Jan 13, 2019)
On Jul 26, 2021 drop support for Numpy 1.17 (initially released on Jul 26, 2019)

What do you think? This could reduce the development burden by reducing the number of older versions we support. On the other hand, we would have to be more proactive about adding new versions to the CI test matrix.