As discussed in the BE senate: https://confluence.camunda.com/display/CO/2021-06-18+Backend+Senate?focusedTaskId=6

Spotify run their new tests x number of times before committing to ensure they aren't committing brittle tests. This has been a problem for us in the past, and it is hard to identify/fix brittle tests after the fact. We should do a mini-spike to investigate whether or not there is any cool tooling around detecting the introduction of brittle tests, and seeing if we can build this into our build pipeline.

In addition, there might be scope to test the running time of new tests too, where we could have upper thresholds for new tests to be introduced to the codebase.

Assigning to andromachi.rozaki as she expressed an interest in this topic as a slack time project. Feel free to reassign to me/unassign if that is no longer the case