As with any software, versioning and deployment are some of the hardest things to get right and with a scaled out scenario with multiple physical locations and multiple target nodes to run software on, this becomes even harder and is very error prone if one has to do it manually.
When working with time series data pipelines and the pipelines consist of multiple different modules that go together and to top it off with then multiple edge locations, this becomes very interesting.
With the Dolittle platform and the continuous improvement engine, all this complexity goes away. This also applies for deploying time series modules, as it is just another piece of software that needs to run somewhere. Working together with our edge support, it has all been optimized to make this seamless for the developer.