Key Notes
- Pavel Rabtsevich used Claude Code to investigate a possible planet in archived TESS data.
- The TESS mission has approved two-minute follow-up observations beginning October 31.
- The signal remains a candidate and has not been confirmed as a planet.
An amateur astronomer has used Claude Code to investigate a possible planet around a star about 116 light-years from Earth, and NASA’s TESS mission has approved follow-up observations. Pavel Rabtsevich’s finding remains a planet candidate, but an official observing program gives the project a route to testing its predictions against fresh telescope data.
The mission’s active list names Rabtsevich as the investigator for program 100, covering two-minute observations of the nearby K-dwarf star TIC 4206066. That confirms the observing allocation. It does not establish that a new planet has been discovered or that NASA has endorsed every conclusion in his analysis.
How Claude Code Helped Investigate the Signal
Rabtsevich, a technology product manager who pursues astronomy as a hobby, described the work in a Reddit post on October 7. He said Claude Code downloaded and parsed TESS archives, wrote analysis scripts, fitted possible transits and produced plots. His role was to choose the questions, decide which checks to run and judge whether their results held up.
The workflow involved Claude Opus 5.5 and Fable 5.1, alongside separate code audits. According to his account, an audit caught an overstated statistical claim, which he removed before rerunning the analysis. He also tested whether a model fitted to two observing years could predict the timing of dips in a third year kept out of that fit.
What the Possible Planet Looks Like in the Data
His September 25 research note reports a signal repeating approximately every 3.18 days across 23 transits. Its depth is about 500 parts per million, equivalent to a 0.05% reduction in the star’s measured brightness. If the object crosses the target star itself, the estimated radius would be about 1.4 times Earth’s.
The paper also describes a weaker signal repeating every 11.13 days, with substantially less confidence. Neither signal is statistically validated, and the author identifies an unresolved companion star as an alternative possible host for the stronger signal. A size estimate conditional on the host star does not establish the object’s mass, composition or suitability for life.
TESS searches for planets through changes in stellar brightness as objects pass in front of their stars. Turning a repeating dip into a reliable discovery requires checking what produced it and where it originated. The candidate label preserves those unanswered questions while allowing other researchers to examine the evidence.
TESS Follow-Up Begins on October 31
Rabtsevich’s October 6 test protocol sets out predictions for Sector 110, scheduled from October 31 to November 26. It fixes nine expected transit times for the stronger signal and three for the weaker one, together with the processing steps and decision rules. The protocol also calls for checking whether the stronger signal’s position is compatible with TIC 4206066.
The distinction from the earlier archival checks matters: the existing data had already been inspected when the signals were selected. The upcoming observations did not exist when the protocol was written, giving the predictions an independent test. Its predefined outcomes include a recovered signal, inconsistent observations, exclusion and insufficient data; a successful recovery would still not prove the object is a planet.
AI-Assisted Science Still Depends on Verification
The project illustrates how a coding assistant can help a motivated researcher build and scrutinize an analysis pipeline. Its strongest feature is the opportunity for a check that could contradict the proposed explanation. Public predictions, supporting code and observations make the claim easier to assess than a persuasive chatbot response alone.
A similar distinction runs through our coverage of AI-generated mathematics papers: producing a promising result and establishing its correctness are separate steps. For TIC 4206066, the immediate milestone is an approved observing program and a testable candidate. Whether it becomes a confirmed planet remains an open scientific question.
Disclaimer: AIstify is an independent media brand owned and operated by NuvexMedia LLC, publishing news, research, and insights on artificial intelligence, emerging technologies, automation, and related industries. NuvexMedia LLC invests in and collaborates with companies across the AI, technology, software, and digital innovation sectors. These relationships do not influence AIstify’s editorial coverage, and the publication maintains full editorial independence to provide accurate, timely, and objective information. © 2026 NuvexMedia LLC. All rights reserved. This content is for informational purposes only and should not be considered legal, tax, investment, financial, or other professional advice.
