follow this procedure

Then if you wish to code with your own specific kernel, build it as a conda environment

Create a user kernel for jupyterhub:

  1. Navigate in your conda environment
  2. conda create -n myconda python=3.7.5 ipykernel

  3. Check your envs

    conda info -e

  4. Upgrade your conda env as you like

    conda install python=3.8.5 # for instance

  5. Inform jupyterhub of new potential kernel 

    .conda/envs/myconda/bin/python -m ipykernel install --user --name 'chris-myconda' --display-name "Chris myconda Env"

  6. Check that the new kernel is there

    jupyter kernelspec list

  7. In case you wish to remove and dereference your kernel:

    jupyter kernelspec uninstall chris-myconda