The Meetup That Helped Build New York’s Open-Source Data Community

For more than a decade, Lander Analytics has helped organize one of New York’s most durable data communities

In 2009, Josh Reich held the first New York R Meetup at the Union Square Ventures office. Twenty-one people RSVP’d. I found the group three months later because Andrew Gelman mentioned it on his blog, and I distinctly remember learning about the head() and tail() functions and wondering why I had made it through graduate school without anyone showing them to me. It was a small thing, but I learned something useful and kept going back.

Drew Conway soon became a co-organizer and started serving pizza, which eventually led to us collecting years of pizza data because of course it did. At one meetup, a group of us were discussing how nice Stack Overflow was and then realized that we had been answering each other’s questions. We knew the usernames before we really knew each other.

That was when the group was still small enough to fit into a small Columbia classroom. Since then, thousands of people have come through the meetup, with speakers and attendees joining from across New York and well beyond it. Along the way, Drew renamed it the New York Open Statistical Programming Meetup to make room for Python, Julia, SQL, Go and the other open-source tools people were already using.

R is still a major part of the meetup, but nobody has to argue that one language should do everything. Most of the people attending use several languages usually because their projects required it.

The meetup became a community

The talks were always important, but a lot happened around them. People found jobs, hired people, met collaborators and made friends. I met my wife Rebecca at a meetup during Michael Kane’s talk about PubMed, then reconnected with her about a year later. The most heavily attended meetup we ever held, when Hadley Wickham spoke in 2015, happened to be our fifth date. We eventually got married, so I am probably biased when I say the social part of the meetup matters.

There are attendees from finance, healthcare, government, media, universities, startups and companies that are difficult to categorize. Some have been writing statistical software for decades. Some are analysts who mostly use SQL. Others are students attending their first technical event in New York and hoping they do not get asked a question. They generally do fine.

The point has never been to make the room feel like an examination. A speaker shares something they have built, researched or spent too much time debugging, then people ask questions. Sometimes the questions are highly technical. Sometimes someone needs a term explained. Both are useful, and an experienced speaker can usually learn something from the second kind too.

More of the community is online now

The meetup had an online presence fairly early. We had the Stack Overflow connections, then a website, Slack, recorded talks and livestreams that let people participate from outside New York. When meeting in person was not possible, those systems kept the group going. We still stream events on YouTube because virtual access is genuinely useful and we don’t want someone to miss a talk simply because they can’t get to the event on a weeknight.

At the same time, there is something useful about being in the room (even beyond the free pizza). People stay after the formal questions, compare notes and discover they are working with the same package, the same strange database problem or in adjacent parts of the same industry. Conversations that might not happen on a scheduled call tend to start naturally.

After the talk, we usually pick a nearby bar and keep going. Some of the conversation stays technical, but not all of it. People talk about jobs, projects, neighborhoods, conferences or whatever else comes up once nobody is standing at the front of the room. The online side lets more people participate, which matters. At the same time, more people seem to be looking for reasons to spend time together IRL, and the meetup gives them one.

The talks change with the work

The subject matter has expanded with the field. Early meetups leaned heavily toward R, statistical computing, visualization and package development. Later we saw more Python, data engineering, machine learning infrastructure and deployment.

Some of those older talks are especially interesting now because the software being demonstrated eventually became ordinary. Wes McKinney spoke about pandas while Python’s data ecosystem was still taking shape. We had talks on Docker, DuckDB, {drake}, {targets} and the modeling tools that became tidymodels, often while the projects were young enough that people in the room could ask the creators fairly basic questions about how they were supposed to work. A lot of us went on to use those tools for years.

More recent speakers have covered public-data packages, machine learning orchestration, MCP’s and (of course) language models. We do not always know which new projects will last, and that is fine. It is useful to hear from someone who has actually built or used the thing while they are still in the room to answer questions.

Why Lander Analytics keeps organizing it

Running the meetup involves finding speakers, scheduling dates, securing a room, handling registration, ordering pizza, setting up the livestream and making sure people know where to go. None of this is particularly glamorous, especially when the audio stops working or a pizza delivery ends up at the wrong entrance.

Lander Analytics has handled much of this work for most of the meetup’s history.

There is an obvious benefit to us. We meet smart people, learn what teams are working on and hear which technical problems keep showing up across different companies. Some attendees later become speakers, employees, collaborators or clients. The meetup also helped develop the community that eventually became the New York R Conference and then the New York Data Science & AI Conference.

Mostly, though, we continue organizing it because we like having the community. The people who attend have taught us a lot, and New York should have places where a new programmer, an experienced statistician and an engineering leader can eat pizza together and have a normal conversation about their work.

Join us in August for a Pixar movies talk

Our August meetup set for August 11 at 7PM EST at NYU Pless Hall features Eric Leung presenting “Data Exploration of Pixar Films: Which One Is the Best?”. This is another good example of the non-traditional talks that have always fit well at the meetup, where an unexpected subject becomes an excuse to look closely at the data and argue about what the results actually mean.

Eric built the {pixarfilms} R package to explore box office performance, compare franchises and see how Pixar films rank across critics and media sources. He will also cover some of the less glamorous work behind the package, including collecting, cleaning and reconciling the data well enough to make those comparisons useful.

There will almost certainly be some disagreement about the rankings, which should make for a good discussion. We’ll have pizza, and some of us will probably keep talking at a nearby bar afterward. If it is your first meetup, that is fine. Most of us showed up not knowing many people either.

Jared P. Lander

Founder and Chief Data Scientist

Lander Analytics

---

Subscribe to our Substack and below to our monthly emails for practical AI strategies for your organization: what to build, what to avoid, and how to make systems reliable in the real world.

Work with us: If you want help identifying the right first workflow, building a permissioned knowledge base, or training your team to ship responsibly, reach out at info@landeranalytics.com.

About the author: Jared P. Lander is Chief Data Scientist and founder of Lander Analytics, where he helps organizations build practical, measurable AI workflows grounded in strong data foundations.