Zerve × ODSC AI: Datathon | Boston/Virtual
The Ultimate Test for Data Practitioners
Transform raw behavioral data from a live AI-native workspace into structured product insights. $10,000 Prize Pool.
Modern applications generate millions of events, but raw logs don’t make decisions - people do. The real challenge is bridging the gap between activity logs and actionable product insight.
In this 24-hour datathon, you won't be playing with "toy" datasets. You will get under the hood of a live AI-native workspace, working with large-scale event data including user interactions, technical metadata, and session activity.
Why Join?
Real Product Data: Authentic signals from a live environment.
Feature Engineering: Design your own signals from raw behavioral data.
Insight-Driven Judging: We value methodology and rigor over "black box" luck.
$10,000 in Prizes: Including cash, Zerve credits, and ODSC speaking slots.
Access AI Expo and ODSC AI East Conference Demo talks - https://odsc.ai/east/
Prerequisite Checklist
This challenge is for intermediate-to-advanced data practitioners. Here's what you should be comfortable with:
Python or R for data manipulation
Pandas / SQL for data wrangling
Feature engineering concepts
Basic statistical reasoning
Jupyter / notebook environments
One-Click Starter Repo
Clone our GitHub template with sample data loaders, notebook scaffolding, and submission helpers — start analyzing in minutes, not hours - https://github.com/
What makes a Datathon different from a Hackathon? While a hackathon focuses on building software, a Datathon focuses on the "Why." You are evaluated on your ability to extract insights, perform rigorous feature engineering, and communicate a clear analytical framework.
Can I participate virtually? Yes! This is a hybrid event. Remote participants have full access to the Zerve cloud platform, digital mentorship, and the submission portal.
What tools will I use? Zerve provides a collaborative cloud notebook environment supporting Python, R, and SQL. You'll receive a One-Click Starter Repo with data loaders and scaffolding to ensure you spend your time analyzing, not configuring.
How are teams structured? You can work solo or in teams of up to 4 members.
Ready to turn raw data into a $10,000 win?
Useful Links
Free access to more talks/trainings: Ai+ Training platform
ODSC blog: https://opendatascience.com/
Slack Channel: https://hubs.li/Q03SyZP80
Code of conduct: https://odsc.ai/code-of-conduct/