Careers at Kaggle

Mission

Kaggle’s mission is to help the world learn from data.

Founding story

In 2010 Anthony Goldbloom and Ben Hammer founded Kaggle, a platform designed to host machine learning competitions, in Australia. It hosted two main types of contests. One enabled organizations with data science problems to provide their data to teams of statisticians who would attempt to produce models from it. The other enabled recruiters to evaluate job candidates for the best fit.

The crowdsourcing approach taken by the company was based on the idea that there are numerous strategies that can be used for any predictive modeling effort, and that it is difficult to know at the beginning which analyst or technique will be the most effective. Based on its significant promise, Kaggle was able to raise $11 million in 2011 from Khosla Ventures and Index Ventures.

The next few years saw a number of milestones. In April 2015 Kaggle released its Scripts product, which enables customers to write, run, and share their code. In September 2015 it raised an additional $1.75 million. In January 2016 it unveiled its Datasets product, which made various datasets on the platform public. In July 2016 Scripts was renamed Kaggle Kernels.

Business model of Kaggle

Customer Segments

Kaggle has a multi-sided business model, with two interdependent customer segments that are both needed in order to operate:

  • Data Providers – Organizations and individuals that provide datasets for competitions. They include everything from corporations to governments to academics to journalists to hobbyists.
  • Data Solvers – Statisticians and data miners who develop models from the datasets.

Value Proposition

Kaggle offers three primary value propositions: accessibility, convenience, and brand/status.

The company creates accessibility by providing a wide variety of options. The datasets featured in its competitions cover every topic from health to government to games to dating trends.

The company offers convenience by making life simpler for customers. It provides a consulting service to help data providers with several competition tasks. These include framing the competition, anonymizing the data, and integrating the model that wins into providers operations. Data solvers’ submissions are scored immediately after entry and presented on a live leaderboard.

A search box on Kaggle’s website enables data solvers to easily find new datasets. Lastly, providers can use its in-browser analytics tool, Kaggle Kernels, to execute, share, and provide comments on code for all open datasets, as well as download datasets in a user-friendly format.

The company has established a strong brand due to its success. It bills itself as the largest community of data scientists in the world, with over 600,000 “Kagglers” from over 194 countries.

The community is active, completing 4,000 forum posts a month and over 3,500 competition submissions a day. Kaggle has held over 200 competitions since being founded, and the contests typically draw more than a thousand teams. Numerous academic papers have been written based on findings from the competitions.

Lastly, the firm’s prominent clients include Allstate, BNP Paribas, Booz Allen Hamilton, Facebook, Genentech, General Electric, The Home Depot, Liberty Mutual, Merck, and Microsoft.

Channels

Kaggle’s main channel is its website. The company promotes its offerings through its social media pages and participation in conferences.

Customer Relationships

Kaggle’s customer relationship is primarily of a personal assistance nature. The company assists data providers by providing them with consultants to help them prepare for competitions. It also provides general e-mail support on its website.

That said, there is a self-service component. Its website provides answers to common questions. There is also a community element in the form of forums.

Key Activities

Kaggle’s business model entails maintaining a common platform between two parties:  data providers and data solvers.

Key Partners

Kaggle partners with organizations to host up to five pro-bono research contests per year. The organizations are of a research, academic, or non-profit nature, and provide a cash prize.

Key Resources

Kaggle’s main resources are its human resources, who include the consultants and other employees that organize competitions and provide advisory services to customers.

As a relatively new startup it has relied on funding from investors, raising $12.75 million from 11 investors as of September 2015.

Cost Structure

Kaggle has a value-driven structure, aiming to provide a premium proposition through significant personal service and frequent service enhancements.

Its biggest cost driver is likely research/development, a fixed cost. Other major drivers are in the areas of customer support/operations and administration, both fixed costs.

Revenue Streams

Kaggle has two revenue streams: 1) revenues it generates from a fee it charges customers to license its platform, and 2) revenues it generates from fees it charges for problem setup consulting.

Our team

Anthony Goldbloom,
Co-Founder and CEO

info: Anthony earned an undergraduate degree in Economics and Econometrics at the University of Melbourne. He previously served as a Research Econometrician at the Reserve Bank of Australia and as an Econometrician at the Australian Treasury.

Ben Hammer,
Co-Founder and Chief Technology Officer

info: Ben earned a B.S.E. in Biomedical Engineering at Duke University. He previously served as a Whitaker Fellow at École Polytechnique Fédérale de Lausanne and as an Associate Engineer at Medtronic.

Jeff Moser,
Chief Architect

info: Jeff earned a Bachelor of Science degree in Computer Science and Mathematics at Purdue University. He previously served as a Senior Software Engineer at Interactive Intelligence and as a Software Engineer at Raytheon Technical Services and Metron.

William Cukierski,
Head of Competitions and Data Scientist

info: William earned a Bachelor’s degree in Physics at Cornell University and a Ph.D. in Biomedical Engineering at Rutgers University. He previously served as a Researcher at The Cancer Institute of New Jersey.