NYC Data #362: CDC at Squarespace, Freezing MMM Coefficients, NYC DOT, Wonder, Working with CRAN, Services for the Underserved, FDNY
Plus, using data to find a seat on the LIRR
Hi friends, happy Friday and L’shana tova umetuka to those who observe: wishing you and your family a happy and sweet 5786. We got to see Mayor Eric Adams rocking quite the Bukharan robe this week so we’re already off to a good start.
As always, help me keep this space up-to-date: please send me posts, events, and job openings. If you know someone who might enjoy or benefit from this newsletter, please share it with them. [image credit: Mario Tama/Getty Images]
Good Local Posts
Squarespace’s own Pravish Sood wrote about our migration to CockroachDB and how we use Change Data Capture. Our work to improve infrastructure here has had a really positive impact on our work as a data organization.
My LinkedIn feed has been full of sophisticated takes on MMMs (nice work, algorithm). Charlie de Thibault exhorts us not to freeze coefficients, which I don’t fully agree with. Low-volume channels are noisy; at some point your testing / holdouts will imply that a channel has a negative impact and applying a strong prior or a limit is better than slavishly following the noise! Elsewhere, Michael Kaminsky talks about how Recast bounds their experimental results over a time period; I think this method is pretty clever.
The MTA wrote a fun data piece about LIRR’s open data on occupancy. Only 1.6% of trips were at least 80% full at some point during their journey in 2024. Inbound trains are least crowded at the front, outbound trains at the rear, consistent with people mostly crowding around the nearest entry/exit point at Penn Station!
Upcoming In-Person Events (new listings in bold)
9/21 - 9/28: Climate Week NYC
9/22: Turning Data into Direction: Shaping Careers in Data Science
9/25: What's the Big Deal with Postgres?
9/29 - 10/3: MLCon
10/6 - 10/12: AI Week New York
10/14 - 10/15: Spatial Data Science Conference
10/22: Introduction to Analysis of Public Survey Data
11/13: Data + AI World Tour
11/13: New York Data Protection & Security Summit
Open Roles
Squarespace is hiring Data Engineers, Senior Data Engineers, a Senior Product Analyst and a Senior Media Analyst.
NYC’s Department of Transportation is looking for a Data Engineer.
FDNY is seeking a Research Scientist to support the Bureau of Fire Prevention.
Wonder is hiring a Senior Data Scientist with an Operations Research focus.
Services for the Underserved is looking for a Data Engineer.
NYC’s Department of Design and Construction is seeking a Lead OMB Data Analyst.
Miscellany
Julie Tibshirani (yes, of the Tibshiranis) posted about her experience publishing an R package on CRAN, and how she appreciates the dependency requirements more now that she’s inside a big, complex company. I’ve never published a package there but I’ve heard friends tell war stories of working through a certain statistician’s feedback. As an end user, I appreciate all the hoops the maintainers go through: it really does result in a better experience!
How Isaac Newton Discovered the Binomial Power Series. From HN: a great study of the process of reasoning by analogy in mathematical investigation!
Thanks so much for being a subscriber. To see previous job listings (many of which are still open!) and blogs, check out the archive, which has emails from the tinyletter days. Feel free to forward this to anyone: they can subscribe here:
