# Starithm NovaTrace — a multi-messenger event tracker built for researchers, looking for input

**URL:** <https://ask.igwn.org/t/starithm-novatrace-a-multi-messenger-event-tracker-built-for-researchers-looking-for-input/1466>\
**Category:** Data Analysis\
**Created:** [April 18, 2026, 8:20am UTC](https://ask.igwn.org/t/starithm-novatrace-a-multi-messenger-event-tracker-built-for-researchers-looking-for-input/1466 "2026-04-18T08:20:19Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![kushasahu](https://yyz1.discourse-cdn.com/flex031/user_avatar/ask.igwn.org/kushasahu/32/579_2.png) [@kushasahu](https://ask.igwn.org/u/kushasahu)\
**Post date:** [April 18, 2026, 8:20am UTC](https://ask.igwn.org/t/starithm-novatrace-a-multi-messenger-event-tracker-built-for-researchers-looking-for-input/1466/1 "2026-04-18T08:20:19Z")

</div>

Hi all,

I’m building **Starithm NovaTrace** ([starithm.ai/novatrace](https://starithm.ai/novatrace/events)), a real-time astronomical event monitoring platform aimed at researchers working in multi-messenger astronomy. I wanted to share it here and ask for honest feedback from people who actually work with these alerts day to day. **What it does right now:** - Consumes live GCN Kafka streams from Swift, Fermi, LIGO/Virgo, IceCube, Einstein Probe, SVOM, CHIME, and DSA-110

- Groups stream notices into canonical events and generates AI summaries from notices + GCN circulars

- Extracts structured measurements from circulars (RA/Dec, fluence, peak flux, spectral parameters, redshift, etc.)

- Interactive 3D sky map showing all recent events by position and type

- Searchable GCN circulars archive  
Example event: [https://starithm.ai/novatrace/events/GBM\_794811435](https://starithm.ai/novatrace/events/GBM_794811435)

**What’s currently in progress:**

- A Bayesian cross-matching pipeline to automatically correlate GW, GRB, neutrino, and optical detections from the same astrophysical source. The idea is to compute false alarm probabilities for spatial/temporal coincidences and flag likely associations across messengers, with LLM-assisted verification of the circulars.

**Questions I’d genuinely like feedback on:**

1. What’s most broken or missing in how you currently track multi-messenger alerts?

2. Would automated cross-match confidence scores between GW candidates and GRB/neutrino detections be useful or is this already well-covered by existing tools?

3. Would anyone like programmatic access to event data and associations?

Thanks