Dan Foreman-Mackey

I'm a Sagan Postdoctoral Fellow at the University of Washington working on next generation astronomical data analysis. My main research interest is the application of probabilistic data analysis techniques to interesting datasets in astronomy. These days, I'm mostly working with Kepler data; everything from the raw pixel values to catalog level inferences. I'm also interested in the development of scientific software and open-source practices.

Publications

My full list of publications is available on ADS but here are a few recent highlights:

D. Foreman-Mackey, B. T. Montet, D. W. Hogg, T. D. Morton, D. Wang, and B. Schölkopf, 2015, A systematic search for transiting planets in the K2 data.
[code, data]
D. Foreman-Mackey, D. W. Hogg, and T. D. Morton, 2014, Exoplanet population inference and the abundance of Earth analogs from noisy, incomplete catalogs
[code, data]
D. Foreman-Mackey, D. W. Hogg, D. Lang, and J. Goodman, 2013, emcee: The MCMC Hammer
[code, docs]

Talks

I post most of my slides to Speaker Deck but here are a few that I really love giving:

Exoplanet population inference — Slides for my talk about Foreman-Mackey, Hogg, & Morton (2014).
An astronomer's introduction to Gaussian Processes — A motivation and introduction to the use of Gaussian Processes in astronomy. The level is aimed at interested astronomers who have experience fitting models to data.
Licenses in the wild — I gave this talk at the 225th meeting of the American Astronomical Society. It is about software licensing and it's based on my analysis of 1.5 million repositories from GitHub.
Hack & Tell (1, 2, 3, and 4) — One of my favorite events in NYC is the Hack & Tell Meetup where local coders, hackers, and makers talk for 5 minutes each about fun projects they've been working on in their spare time. I've presented at this meetup 4 times so far and it's always been a blast.

Code

I write a lot of code for work and in my spare time. All my projects live in public repositories on GitHub. Here are some of my most popular research codes:

emcee — Kick-ass MCMC sampling in Python. See the paper.
George — Blazingly fast Gaussian processes for regression. Implemented in C++ with Python bindings.
triangle.py — Simple corner plots (or scatterplot matrices) in matplotlib.
Daft — Pixel-perfect probabilistic graphical models using matplotlib. (Pretty much unmaintained at the moment but hopefully that will change some day).