Getting Started

If you are a student interested in working as part of vault this page will help you get started.

Connecting to the Community

Before you get started on any of the fun stuff, first you should connect to the rest of the vault community. We share our interesting findings, help each other when we get stuck, and just provide a place to chat.


Join our Slack Workspace!
We communicate through Slack, a communication tool widely used by professionals at college-level and beyond.
You should prioritize signing up for our Slack Workspace, because if you have trouble this is one of the main ways of getting help.


Find us on GitHub!
We share our work through GitHub, an industry-standard collaboration tool.
If you do not yet have a GitHub account, sign up for one.
If you would like to be part of our GitHub organization and have not gotten an invite, contact us.

Finding a Question

Once you are set up to talk with your peers, you can move on to choosing a question: something you would like to understand about the world. This is both the hardest and the easiest part. Some example questions are “Are crimes more likely to occur on a particular day of the week?”, “How different are top-ranked colleges from one another?”, and “Are there undiscovered objects in the famous Hubble Deep Field image?”. There is no perfect way to come up with a question, but we list two methods immediately below. If you already have a question, you can move on to the next section.

Data First, Question Second

In many cases questions stem from having an experiment, data, or a theory in front of you. This is frequently what happens in university-level research. For this, download or browse through a data set (and you can find many of those on our resources page), and see if any questions pop out to you. Many of the previous projects came from doing just that. You might even just look at your favorite sci-fi or fantasy movie, and try to understand the physics and mathematics behind it.

Pre-Existing Questions

Another fine way to come up with a question is to use one that already exists. Check the projects on our github, for instance. While they are interesting explorations, most of them have a number of unanswered questions you could make your own. There are also a number of science/math competitions available, some of which come with pre-assigned problems, for example the Mathematical and Interdisciplinary Contests in Modeling Also check out this discussion on the blog of a well-known Social Sciences/Math Professor about potential projects for High School Students.

Learning to Answer a Question

Once you have a direction you would like to go, the next step is to start your exploration! In this case, your next step depends on your question. For example, if you have decided to use a pre-existing project as a starting point, you can open up the analysis notebook in Colaboratory and fiddle with the data. If you’re lost, consult with your peers and mentors on our Slack on the next steps to take.

As a final note, persistence is your most valuable tool. If your exploration requires programming and you don’t know how to program, look up some programming lessons. If you hit a dead-end, try something different for a while, or ask a peer or mentor. A large part of a successful data expedition is not giving up and not letting yourself get stuck on one aspect of the analysis.