Today I spent the day at the Microsoft offices near Times Square participating in an open source sprint for the scikit-learn Python library organized by the NYC chapters of Women in Machine Learning and Data Science and PyLadies. While I'd done hackathons before, I'd never participated in an open source sprint, so here are some of the takeaways from the day.

Things I learned about Open Source Sprints:

• Prepare before the sprint! The sprint had an entire github repo with all the information we needed to get started, which was key for avoiding wasting an hour configuring my particular Python installation. I ended up creating a new virtual environment to keep my "work" scikit-learn installation separate from my "development" installation.
• Working with a partner holds you accountable. I worked with @olgadk7
• Ask questions! We were fortunate to have several of the scikit-learn core contributors, including Andreas Mueller, Thomas Fan, and Nicholas Hug on deck all day
• Start small. We worked on one minor documentation fix before tackling a larger example documentation fix.
• If possible, find a similar issue that has been fixed to see what the standard is for that kind of issue.
• Tell people what you're working on! If you don't comment on the thread, especially in a sprint where people are looking at the same issues, you might get scooped!

This is the third year that WiMLDS has run this sprint and it showed in how well everything was organized and set up. I was very happy I was able to participate, as spots were limited based on the space we were in at the Microsoft offices! Our minor documentation fix has already been approved, and the second documentation one is in progress.