[UX Design] Improve/iterate on Hamilton execution view
- Dominant language
- Jupyter Notebook
- Stars
- 2.6k
- Forks
- 213
- PR merge metrics
- No merged PRs in 30d
Description
The Hamilton execution view contains a lot of information, but I feel we could surface things more clearly. Currently things are very separate:
In this view at the top we have "parameterization" of the DAG that was run. This could expand to have other metadta.
In the middle there's a "dataflow" view that you can switch for a "waterfall view" of how long things took.
At the bottom lists what was execute where you can click for more details.

Once you click in you come to these three views -- a data summary -- what's shown here will change based on the data type.

A view of the code:

A view of the errors (if any)

Things I think could be a better UX:
- Hamilton is inherently graphical -- so can we navigate all this execution information by see it on / with the graph? Maybe it has to be vertically oriented by time to fit on a screen? or?
# Task
Come up with some designs.
# Inspiration
Things to take inspiration from:
- https://langfuse.com/
- https://datadog.com/
- https://www.elastic.co/kibana
Contributor guide
Assessment
This issue has not been assessed yet.