Join the Datathon@LISH

Presented by the Laboratory for Innovation Science at Harvard

Do You Have What It Takes to Win a Harvard Datathon?

The Datathon@LISH is a remote two-day data science competition running on the weekend of February 12th, 2022. The event is designed for any US-based university students or affiliates and is organized by the Laboratory for Innovation Science at Harvard.

Grow your data science skills

Hone your data science skills by solving our selected prediction problem for social good. Our selected contest problem is sourced from a non-profit working in global development.

Win cash prizes, swag, and recognition

Cash prizes and recognition among the Harvard data science community for winners. First Place: $1000, Second Place: $500, Third Place: $250, Top-10 Finish: $50. Participation Prize (t-shirt or $10 gift card) for eligible teams that beat the baseline score in the contest.

Help the science of
(data) science

To further knowledge on data science, contest code will be collected for a study on data science problem-solving. Participants will also complete two short surveys before and after the contest.

Competition Details

Prediction Problem

Our contest features a statistical prediction problem similar to ones on Kaggle, AICrowd, or TopCoder. The winner will be the model with the best prediction score on our holdout sample.


Weekend of February 12th, 2022. Exact start and end times are to-be-determined, but contest will be open for about 48 hours (on the shorter end for datathons)

Expected Time Needed to Participate

Work time is flexible and self-determined during the contest. We recommend blocking off at least 8 hours to work on the problem in order to be competitive.

Eligibility + Python Requirement

Any US university affiliates may join (including non-Harvard affiliates). This includes undergraduate and graduate students, recent graduates, and other staff (e.g. RAs, post-docs, etc.). Python programming is required for the contest. Learn more.


Individuals or pairs entry allowed.

Remote Contest

Contest will be run entirely virtually due to Covid restrictions.



Iav, HBS Professor

Paul, HBS Researcher