I am the first designer to work at MAST, a data archive for NASA space telescopes. The archive was composed of many search forms that looked like they were built in the 90’s by scientists. (They were, and they were!) These forms had different design patterns, both visually and for completing tasks—there was much unnecessary complexity. Science workflows are often complex, but they shouldn’t be made even more complex! Over several years, I have designed a system of cohesive components to use across MAST’s many searches. I have established common workflows that are repeated in different contexts. Now, MAST searches are a part of an interconnected system. Scientists who are familiar with one can easily swap to another, making them easier to learn and use. I continue to expand and interconnect this system as new telescope missions are added to our archive. This has been a complex, multi-year project, involving many stakeholders: developers, project scientists of different backgrounds, mission leads, and the astronomers who use MAST to conduct their research. I have stayed focused on our core users: research astronomers. I have led design workshops and gone through iterative rounds of feedback with scientists. I've run usability tests, created surveys, reviewed helpdesk requests, and even tried experimental methods like interactive conference posters.

Jdaviz is astronomical data analysis and visualization tools based on the Jupyter platform. These add a UI interactive layer to computing notebooks. They are designed to work within a Jupyter notebook cell, as a standalone desktop application, or as embedded windows within a website.

For this project I have consulted with the team of developers and scientists on adding features and streamlining existing features. I have designed logos and icons and led usability feedback sessions.

With the extremely high volume of data coming from telescopes like Roman, more scientists will need to use Jupyter environments. We built a series of interactive widgets for Science Notebooks. These widgets help to give science notebooks more non-code interactive options like a skyviewer and Jdaviz tooling, and they combine with python in powerful ways. This enables cool new research workflows that were not possible before.

For this project I advised on UX, ran usability feedback sessions, and provided icons.

Jupyter Notebooks play a central role in modern data science workflows. Despite their importance, these notebooks are inaccessible to people who rely on assistive technology like screenreaders. Impacted users must find extreme workarounds or give up using them entirely. Students with disabilities have reported leaving their field once they learn their chosen career’s foundational tools are inaccessible to them. I led a multidisciplinary team to try to solve this problem.



I gathered a team of people with different backgrounds and skills. We wrote a killer grant proposal and was awarded and administered a 6-figure grant. Notebook accessibility is a super thorny problem and we did not solve it, but we did make progress!

My part of the project was collaboratively writing and conducting usability tests with people who use assistive technology. We did two rounds of usability testing on Content types, structure and navigation. The scripts and results are all publicly available.

Based on the feedback we gathered, our developers experimented with an improved notebook html output. Our lead developer Tony Fast has continued working on this in his project nbconvert-a11y. We also established best practices for writing notebooks accessibly. We condensed this information into an accessibility checklist that people building notebooks can reference.

As part of this grant we organized public events: Accessible Notebook Hackathon and Day of Accessibility, and presented our results at Jupytercon 2023 to try to gain awareness and get more people working on this problem.

Astronify is a python package that sonifies--turns to sound--astronomical data. I was a core member of a small team awarded a grant to pursue this experimental work.

I led the usability testing of the sonifications. This was an adventure through uncharted jungle, exploring user-patterns that aren’t well established. We focused on how sonification could be useful for scientific research—they needed to be interactive and convey meaning that could be used by a scientist, not simply tell a story. This is fundamentally different than a curated sonification for the general public. I worked with GLAS education to connect with blind and low vision technologists who I met in one-on-one sessions to gather feedback. I also used gamified surveys to gather asyncerous feedback. People play a game trying to identify meanings within sonifications, and based on their answers, we learned if the information in the sonification was discernable.



I also led community outreach. I created a series of videos explaining sonification and astronomy concepts to an amateur audience. For these videos I wrote and recorded the script, edited the recordings, came up with a visual language that would be the most visible for low vision audience, and animated the videos. The videos were featured as a learning resource on the Perkins School for the Blind website. I connected with the NPR show Out of the Blocks which featured Astronify. I presented the an astronify gameshow at a sonification festival and conferences internationally.

I collaborated with Iota School to organize this event. We gathered people working in science and tech to engage with the challenges around making science and tech inclusive. Through a series of talks and hands on learning activities. Attendees learned about accessibility challenges in STEM, and will left prepared to act as accessibility advocates in their work. All talks and slides are publicly available.

Our keynote was given by Joshua Miele, a MacArthur fellow and leader in adaptive technology design. Science writers talked about best practices in astronomy alt text, the notebooks for all team spoke about Jupyter Notebook accessibility, and Dr Scott Fleming spoke about data sonification. Dr Patrick Smyth of Iota School a hands-on workshop to learn the fundamentals of using screen readers. We had interactive tactile exhibits on display. We were pleased to welcome a diverse audience including a large cohort from the Baltimore School for the Blind.

I collaborated with Iota School to organize this interactive hack event for Jupyter Notebook authors. We started with a talk explaining accessibility issues with notebooks—they don't read well on screenreaders—and practical tips for making notebooks more accessible. Then, atendees spent an hour applying what they learned in a notebook they were working on. We ended by gathering as a group and sharing our results.

I was invited to run a similar event at Rubin Observatory.

I envisioned, found funding, and organized this paid internship program for high school artists. Each student creates art inspired by astronomy data in the medium of their choice. We had students create paintings, sculptures, music, animations, and augmented reality apps. I collaborated with colleagues to run the program.

In 2023 We hired a cohort of 12 interns from Baltimore to create art inspired by astronomy data. The program was 6-weeks and the interns worked 10-15 hours a week. We met in person at Space Telescope Science Institute once a week for 2.5 hours. Alongside time creating artworks, the students met MAST scientists, wrote a weekly progress journal, participated in educational talks, and visited James Webb mission control. We ended with a gallery show at STScI which was widely attended and well enjoyed.

In 2021 We hired students to create art inspired by space sonification that was accessible by a blind audience. We met by video chat for 6 weeks. The student artists were both blind and sighted themselves—so we had a lot of interesting discussions about accessibility and art.

1 / 11

Acrylic paintings of MAST data on display

2 / 11

A spiral galaxy made from broken glass

3 / 11

A lamp whose shape was inspired by MAST data

4 / 11

A diorama of the cosmic web, with space spider

5 / 11

A student shows off their AR piece

6 / 11

A fabric sculpture

7 / 11

A visiter touches a 3D painting

8 / 11

A geometric lamp surrounded by onlookers

9 / 11

Students show off their work

10 / 11

A student explains his music to a father and young daughter

11 / 11

A student explains his data website to interested onlookers

Using historic census data from 1790 - 2020, I explored data Baltimore population through sonification. The final recording and the notebook I used to generate the files are available on Github.

In Baltimore there is palpable tension between the city and the county. Even the name. Baltimore county refers to itself as Baltimore and to the city as Baltimore city. City people call the city Baltimore and Baltimore County is just the county. (With a air of distaste on both sides) They fight over the name Baltimore itself.

I started by visualizing the information on a graph using Excel to find out if I could see the story before I listened for one. I only had data until 2010, so I found the most recent census to complete the set. It was challenging to find that, for a minute I thought I would have to come up with the number by adding all the populations of different jurisdictions in the county, but luckily I found the data I was looking for. I used Astronify to generate the sonifications. It is meant to be used with lightcurves, but I knew it would work as long as I had a list of numbers.

I created a Jupyter notebook and pulled in the census data I created a table that Astronify could use to sonify. Then I generated separate sonifications representing the city and the county’s population data. I could hear something that I hadn’t noticed when inspecting the data visually. There were 3 main sections of differing trends.

Since the sonification by itself is not self-explanatory, I decided to write a script and use that as an opportunity to focus on those 3 section trends. I recorded a few takes of myself reading the script. Then edited the recording and sonification together in Audacity.