What do you like best?
Zeppelin provides a unified notebook to collaborate data science/data analysis methods and techniques. It powers visualization to big data analytics, especially Apache Spark. On the team level, each team member collaborates his/her analytics in one shared workspace. On the individual level, Zeppelin utilizes multiple data analytics technologies in one notebook! For instance, one paragraph could be Scala-Spark code that processes the data in a specific way, which can be visualized in another paragraph using R.
Additionally, Zeppelin provides very interactive visualization with Big Data.
What do you dislike?
Zeppelin has some UI bugs. For example, some of the graphs appear and disappear in random basis while loading very large data or many users are accessing the same notebook at the same time. Also, sometimes, adding new paragraph does not work immediately. The browser should be refreshed to find the newly added paragraphs in the notebook.
Also, the data filteration does not work correctly.
What problems are you solving with the product? What benefits have you realized?
It communicates the data analytics in a very interactive way to the business users. Also, it makes the collaboration between the team members easier. It saves a lot of time compared with other notebook software, where each team member should push the notebook to a source control (e.g. GIT).