This article is featured in the free magazine "Data Science in Production - Download here
At GoDataDriven we have an Open Source First approach. Unless some shockingly good reasons exists,
we always advise to use (and implement) open source solutions.
It is therefore only natural that we have the tendency to give back to the open source community.
Some of these efforts have a good visibility (such as Divolte), while other remain in the shadow.
I therefore thought that I can start to change that by publishing,
every once in a while, the various contributions to open source projects, both old and new.
This first edition starts with Fokko that
contributed to 4 different projects: Druid, Docker-Druid, Airflow, and Flink.
- In Druid PR 3481 he fixed the INFO message logs;
- In Docker-Druid PR 30 he expanded on the logging section in the README;
- In Airflow PR 2038 he resolved session leakage;
- In Airflow PR 2042 (still open) he added a spark-submit operator/hook;
- In Flink PR 3077 he implemented Stochastic Outlier Selection (!);
- In Flink PR 3081 he cleaned up the Flink Machine Learning library (!!).
Next up is Vincent who create a whole new project, Kadro which is a friendly pandas wrapper
with a more composable grammar support. The goal of the library is to have a minimal wrapper that
allows most of all dataframe operations to be more expressive by being chainable.
Then Bas contributed Prometheus JMX exporter PR 108 (still open) to add HBase example config.
Finally, yours truly contributed NiFi PR 1467 (still open) to add the exception class to
failing flow files coming out of InvokeHTTP.
That's all for the first edition. As always, we're hiring Data Scientists and Data Engineers. Head
up to our career page if you're interested. You
get plenty of opportunities to give back to the community.
We are hiring