Software Developer in Truckee, CA, United States

At Nuna Health, as part of the data processing pipeline, we generated a lot of data quality metrics and validations. The problem was that there were so many metrics and validations that it got out of control for a human to comprehend all those generated numbers anymore, so this situation created a lot of manual work.

I started designing a data quality visualization dashboard that would make it easier for users to understand better and monitor the quality of the data.

The design process involved user interviews to better understand the project requirements, a couple of brainstorming sessions and multiple prototypes.

I chose to implement the dashboard using Looker, a business intelligence tool that has nice visualization capabilities for big data analysis. Using Looker's own data modeling language (LookML) I implemented a set of dashboards around the overall quality of the data of a client run, comparison between multiple data processing runs and graphs that highlighted spikes in changes of the data quality over time.

This tool was used by data operator users and eliminated tens of hours of manual work per month.

When I worked at a healthcare startup, I helped design and develop a RESTful API for matching a patient with the best available doctor given a set of constraints like health condition, physical location, age, and more.

It was a very impactful project, shipped to health insurance providers to improve the outcome of patients health care.

Technologies: Python, Django

During my time at Nuna Health, I designed and implemented a system for client data schedule management and notifications around important data events.

This system kept track of the expected schedule for receiving client data, late data updates, expected SLA deliverable dates. I built a web page to display all this information in a user-friendly visual way and a notification system which alerted users via email for important events like late data, the moment when all of the monthly data for a specific client has arrived, and so on.

This project was developed using Python/Django and Angular and helped reduce a lot of manual work by data operator users who were keeping track of all this information manually in an Excel spreadsheet. It also helped avoiding missing important deadlines.

For one and a half years, I had the unique opportunity to be one of the HPE's contributors to the open source project OpenStack. I contributed to multiple parts of the OpenStack repository, and it gave me the opportunity to interact with highly talented developers from other companies which were contributing to OpenStack (IBM, Red Hat, Rackspace, and more).