Interested in joining the operations team?

Behind the BinderHub deployment at is a team of volunteers that are
help to run the service, do general maintenance, run upgrades, and manage infrastructure. For more information about the team and what it does, see the operations team guide.

If you’re interested in helping out with operations, please post here! Please share a bit of information about your background (technical and otherwise), interests, and the ways in which you’d like to help.


I am interested. I have been doing server tasks and upgrades for a few website projects, usually VMs on DigitalOcean. Recently, I have focused more on full-stack application development and some Python data-oriented tasks at work.


hey @brylie! as a start, maybe you can give a shot at bumping versions for binderhub/repo2docker? That’s often a good first-step at getting involved. Check out the deployment guide for some information on how to do this!

1 Like

We don’t have a good on-boarding guide (yet), so bear with us as we figure this out together.

I think what Chris said makes sense. All the “work” is done by travis when we make changes to the deployment repo so there aren’t many commands to type into a shell (most of the time). To get a feeling for what the moving parts of the deployment are I think it would make sense to setup locally on a minikube instance. Don’t know if this is the best way to learn but it would be a way :smiley:

There is also The Operators (no Binder isn't forming a rock band) which gives an idea what happens day to day.



I’d also be happy to help out. I’m deploying a Jupyterhub setup for the Human Brain Project’s Collaboratory (an infrastructure for neuroscience research in Europe). I’m available with operations and development.



I think the best way to get involved at the moment is to start hanging out on and watching

Things are pretty stable at the moment so it is all about optimising and tuning things. For example thinking about why on one of our clusters we seem to cordon nodes a lot more often than on the others. My guess is that we clean out the local docker cache too often with

Hey! :wave: I’d also suggest joining in on one of the team meetings to have a chance to see what folks are talking about there