My Work

I led the team responsible for an overhaul of the Dagster+ UI to make it a central hub for data pipeline observability. This is a video I produced and edited that previewed all the exciting new features we shipped.

As part of our Summer update, I designed an all new homepage for Dagster+ that gives data teams a clear view of their platform, allowing them to triage issues in real time and customize it to their needs. Read the blog post »

I led a redesign of our data lineage UI that allows data practitioners to customize the metadata that appears on each graph node. This makes the graph much more useful for debugging and understanding failures. Read the blog post »

We recently shipped a new feature in Dagster+ called Realtime Insights Dashboards. I let the design and product work to deliver this feature. It allows data team leaders to build flexible KPI dashboards to monitor critical metrics like Success Rate and time to resolution across all the teams and data products they manage. Read the blog post »

This is an unused concept for a revamped execution timeline UI. I was inspired by seeing some of the challenges our users faced with our existing timeline view, particularly how events are bucketed on the timeline. I explored a new approached that bucketed events into 5 minute increments. This allows for a cleaner visualization but comes with a tradeoff that it is less precise. We ultimately decided to abandon this approach.

In this video I walk through a bunch of UI improvements we made to our Open Source version of Dagster. I led the redesign of all the features here.

I produced a vibrant motion graphic for the Dagster Launch Week event. This was used throughout the launch week along with a collection of presentations and webinars that I also helped to produce. This graphic was built in After Effects and edited together in Adobe Premier.

I worked with the Hyperscience team to redesign and rebuild the layout editor UI. This UI is a critical component to their AI document processing software because it allows non-technical users to train a ML on any document (for example, a health insurance claim or employment verification document)

At Hyperscience, I led the redesign of a key developer workflow – debugging custom automation flows. I helped simplify and improve the debugging experience while developers are building and testing their custom workflows.