Kaggle is the world’s largest community of data scientists. It also happens to be one hell of a sourcing tool, helping you to identify one of the industries most difficult to fill job families, Data Scientists. I’d highly recommend browsing through the site to learn more, specifically the competition page to see just how real the rewards are for their user base.
What is Kaggle?
To date Kaggle has racked up just under 300,000 users from more than 100 countries and over 200 global universities. As shared on their site, Kaggle is the world’s largest community of really smart people, we call data scientists. Member’s compete with each other to solve complex data science problems, and the top competitors are invited to work on the most interesting and sensitive business problems from some of the world’s biggest companies through Masters competitions.
Kaggle functions like an open source hack-a-thon type of environment were organizations submit real world challenges that otherwise go unsolved in their current environment. The Kaggler’s (members of the Kaggle community) work to solve the challenge in competition format.
Is Kaggle a Viable Sourcing Ground?
Yes, it is. Kaggle members are real. They are competing with each other for real prize money (sometime exceeding $100K) and of course bragging rights. The challenges are sponsored (paid) by the companies and are most likely related to your clients internal struggles, especially if they are hiring for a statistician or a data scientist.
Users are allocated points for their performance in the competitions that they participate in. All users are than stack ranked globally. You can read about the Kaggle ranking system here
Here are 2 great ways to use Kaggle to recruit top data scientist talent:
Recruit like Facebook:
Facebook has caught on by hosting competitions to solve real world challenges using Kaggle. The competition requires the user to submit a formal solution to the Facebook challenge (which serves as their application, made public and known to the user). The submissions are reviewed and recruiting commences.
The return for Facebook? Through a 3 phase competition, Facebook saw 7,092 entries and of course an onslaught of hires.
You can see the Facebook case study here
And of course the more realistic way that most recruiters without any significant budget will use Kaggle: The good ol’ X-ray / site search:
Step 1: We need to find the correct path for sourcing Kaggle, targeting their user list. You can see from the url in the image below, the structure we need to follow in order to pull user profiles.
Step 2: We need to create our basic x-Ray search. There is little to no need to get crazy detailed in your search.
site:kaggle.com/users/* United States Data -> when we run this search in Google you will see this returned set of results.
Step 3: As you click through to view the user profile you will see that you have access to a number of key areas highlighted in red in the below image.
And that’s all she wrote…
Check out some of their team bios:
Kaggle Scientists: http://www.kaggle.com/solutions/scientists
The Kaggle Team: http://www.kaggle.com/careers/team
Kaggle Users and rankings: http://www.kaggle.com/users
How Users are ranked in Kaggle: https://www.kaggle.com/wiki/UserRankingAndTierSystem