Hi! I’m Drew Levitt. Welcome to Sourdough Data.
About You
Maybe you know me personally; maybe we worked together once. Maybe you’ve heard of me from a colleague or friend. Maybe you happened to google an admittedly unlikely pair of words. Probably, you’re coming from the world of transportation or city planning. But one thing is for certain: you have a question.
About Me
I help people figure stuff out.
I’m a transportation planner by training and a data scientist by experience. I completed a Master of City Planning degree at UC Berkeley in 2016, and worked at Fehr & Peers and the Metropolitan Transportation Commission before setting out on my own.
I believe the fundamental question of quantitative analysis is: “Compared to what?” I am deeply committed to framing information accurately, honestly, and insightfully. I read a lot of Edward Tufte books.
I believe that the systems we build must improve the general welfare, and that urban planners have a professional and ethical responsibility to recognize, acknowledge, and redress the inequitable harms perpetrated by the planners of yesteryear.
I am intrigued by novel questions and datasets; undaunted by new technical frontiers; multidisciplinary in my background and approach; and committed to excellence and innovation in my work products and process.
When I’m not at work, I spend a fair bit of time riding bikes, playing with my cats, and yes, baking sourdough bread.
A few things I’ve done recently:
- At Fehr & Peers, I led a first-of-its-kind study on the relationships between bike/ped infrastructure and emergency vehicle travel speeds. We used a variety of innovative spatial-statistical methods to make sense of a lot of complex, contradictory data.
- Also at Fehr & Peers, I built an interactive web app (in Svelte, a JavaScript framework) that let users forecast the impacts of large-scale trends on vehicle travel and transit ridership.
- At MTC, I helped wrangle large regionwide land use and travel forecasting models (and Python-based pre- and post-processing tools) to accurately ask and answer questions of regional policy importance.
- I developed a Python boot camp to help teach UC Berkeley planning graduate students the fundamentals of data engineering, analysis, and visualization in
pandas. - I’m currently helping MTC’s data team streamline the codebase behind Vital Signs, a regional data dashboard.
Some things I can help you with:
- Python code development and review
- Geospatial analysis and visualization
- Exploratory data analysis
- Production-ready data processing pipelines
- Interactive data visualizations
- Statistical analyses (and, equally importantly, explaining statistical analyses to lay audiences)
- Big data analysis
- Web app development
- Research study design and documentation
- Transit (GTFS) data analysis and storytelling
- Obscure literary references
About Us
Whether you’re building something new and remarkable, or have tough, important questions to answer, I’d love to help. Drop me a line at [email protected] and let’s take it from there.
The mind that is not baffled is not employed.
– Wendell Berry, “Our Real Work”