Rubin Alerts & AI Hackathon 2026

Our Purpose

The Rubin Alert Stream

Preparing the community to harness the most powerful time-domain survey ever built

The Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) will scan the entire southern sky every few nights, generating roughly 10 million difference-image alerts every single night of operations. Each alert carries photometry, cutout images, and contextual metadata for a transient, variable, or moving object. No human can review this flood of data — AI and machine learning are not optional. They are essential infrastructure for doing science with Rubin.

The NSF-Simons SkAI Institute is committed to serving as a community hub, bringing together astrophysicists and AI experts to tackle exactly this challenge. This hackathon, supported by the LSST Discovery Alliance, is a focused effort to develop the tools, models, and pipelines needed to enable early science with the Rubin alert stream — and to build the collaborative community that will carry that work forward into the LSST era.

We look forward to welcoming you to Chicago and the SkAI Institute for three days of focused, hands-on work at the frontier of AI-enabled astrophysics.

Vera C. Rubin Observatory with star trails

What We'll Build

Hackathon Goals

Three days of focused, collaborative development on AI tools for the Rubin alert ecosystem

Real-Time Alert Classification

Develop and benchmark AI models for rapid classification of Rubin alerts into transient types, variable stars, and other astrophysical phenomena.

Alert Filtering & Prioritization

Build intelligent filtering systems that help scientists identify the most scientifically interesting alerts from the nightly stream for follow-up observations.

Multi-Modal Data Fusion

Combine photometric light curves, difference-image cutouts, host galaxy information, and contextual metadata to improve classification accuracy.

Anomaly Detection

Identify unusual or previously unknown classes of objects in alert streams — the unexpected discoveries that make large surveys transformative.

Scalable Pipelines

Design and prototype open-source pipelines capable of ingesting, processing, and responding to Rubin-scale alert volumes in near real-time.

Community Tools & Standards

Foster the development of shared datasets, evaluation benchmarks, and best practices that will serve the broader Rubin science community.

Mark Your Calendar

Important Dates

Mar 3

Registration Opens

Online registration portal opens for all participants. Space is limited — early registration is encouraged.

Mar 20

Registration Closes

Final deadline for hackathon registration. Late registrations may be accepted on a space-available basis.

Apr 1–3

Rubin Alerts & AI Hackathon

Three days of collaborative hacking, presentations, and working sessions at SkAI Institute, Chicago.