Being a tad bit of a perfectionist, I realized that I have a tendency not to put up a lot of work I have done because there are things about it I would change. But, because of school, time to perfect things rarely exists right now, and I'm still in the process of learning and growing. So, I decided I should put some work up and tell you all about what I would change about it given more time, rather than waiting until it's perfect. Here are some of the things I have worked on over the last few years.
A project from my data viz class. The challenge was to build an accurate and inventive clock using p5.js. I chose to invoke the anxiety associated with climate change with my doomsday clock. Given more time, I would build this out further to include facts about climate change with every drop, and add markings to make the clock aspect more readable.
The final project for my advanced topics in databases class: a research paper that analyzed various performance optimizations for performing in-memory query perturbation with Pandas dataframes. Query perturbation is the idea of having a query template on a database, and varying a certain set of parameters to search for a particular result in your database! While the features described in the paper were implemented, what I would improve in the future is the organization of the code, and make it a fully functioning system with documentation that everyday Pandas users could install and actually use.
A conference talk I gave at the o11ycon Observability conference which was centered around how to detect bugs and gain insight into issues your system was having. I worked at a hardware company at the time, so I gave a talk on how to debug systems when the bug could be a software problem, a hardware problem, or both!
Two episodes of a podcast created for our Audio class. In this episode, I played the role of the executive producer, and had an audio feature story. One thing I would change given more time is the recording of my narration, which due to Covid, I couldn't access the sound booth at school! But other than that I had a great time making this podcast with my classmates!
Two episodes of a podcast created for our Audio class. In this episode, I played the role of the host, and hosted a two way interview. I would edit down the host interview in the future, but I learned a lot during this process!
Wrote Python scripts to analyze Trump's tweets for this piece about his interactions with the media on Twitter. Given more time, I would have built out more data visualizations to go with the piece!
A group project for Dataviz. We grabbed NYC's public transportation turnstile data, and processed, cleaned, and visualized the data in a multitude of ways to understand which neighborhoods were able to stay at home when the pandemic began, and which neighborhoods were still forced to go to work. We further compared this to the average income in various neighborhoods by joining our turnstile data with census data in order to understand the economic trends. The whole site is less polished than I would want it to be. Some of the visuals are heavy and load slowly, and could be scaled down in order to have a faster and cleaner page load. Initially we were trying to create a smooth scrolling effect, but ran out of time to do so. Some of the visuals also take too long as time progresses, and should be sped up. All are fixable problems given more time!
A group project for Dataviz to practice d3. We grabbed the 2018 Congressional Race SEC filings to try to understand how much spending on the race correlated with winning. Knowing what I know about D3 now, I would have totally restructured the code, and perhaps tried a different visualization to better show the effects of spending on the chances of winning. But it was fun to practice implementing pop-ups and the clicking behavior in d3.
Podcast epsiode: Produced, wrote, edited and reported two episodes on the Oakland teachers’ strike that interviewed teachers and activists. Learned a lot during my audio class last semester about everything I did wrong when I recorded this, so I would do over the episode if I could!
A quick data challenge that involved mapping HHS data to build an animation and an interactive data visualization that shows the changes and statistics of ICU bed availablity through time.
A guest lecture I prepared and presented for the Advanced Software Engineering class at Columbia about good testing practices in production settings, and how to avoid bugs!