Cool discovery, but I'd be real curious about the false-positive rate when letting coding agents loose on raw astronomical telemetry. Hallucinations in data reduction are no joke.

From the discussion:

> even got NASA to point a telescope at it for confirmation

with a proper stats harness the false positive rate can be brought well down to acceptable levels

[flagged]

If you can set the criteria for testing that a human would also use, then any llm can do the job.

What is uncool is how people are using cloud based AI and their universities are not providing self hosted models that keep data within the university. Cloud AI is useless due to it being no different than publishing everything you do publicly.