Software developer Will Depue and a small team used AI agents over a few weeks to comb public astronomy archives, identifying asteroid and dwarf-planet candidates from nightly telescope data. The MTS podcast reported that the project began as a search for the hypothetical Planet 9 before drifting into unexamined sky data that people lack the time to fully review.
Searching unexamined sky data
Government-funded telescopes, both ground- and space-based, photograph the sky every night and publish the results online. Most of these surveys focus on specific targets, leaving vast amounts of archival data unexamined for other objects passing through the frame.
Will Depue, Luke Farritor, and Josh Wall used Claude and Codex agents to comb through these public astronomy archives over a period of a few weeks.
“There’s just tons of this data out there and a lot of it hasn’t been totally deeply finely combed through because people just don’t have the time to look for every possible object that appears in any of these data sets,” Depue told BigGo Finance.
The project initially aimed to find the hypothetical ninth planet, known as Planet 9. Instead, the agents surfaced asteroid and dwarf-planet candidates, alongside other unidentified objects in the archives.
Institutional blind spots in astronomy
Professional astronomy institutions often miss these smaller solar-system objects because their talent and resources are directed toward higher-priority targets. Depue said that while finding small objects is engaging, institutions do not optimize their workflows for the long tail of archival data.
The primary barrier to searching this data is human attention rather than data access.
With the cost of software engineering and data analysis falling, abundant compute power allows small teams to process information that larger organizations leave behind.
“There's a massive number of unimportant results they didn't publish or that people independently are discovering,” software developer Will Depue argued that headline science captures only a sliver of what public datasets contain. He pointed out that institutions frequently publish only their most prominent results, leaving a massive number of smaller discoveries to independent researchers.
Funding whimsical research with Macro Grants
To support independent projects, Depue operates Macro Grants, a nonprofit aimed at whimsical research that falls outside conventional grant categories. The organization provides grants ranging from $500 to $25,000, delivering decisions within two to three days.
Alongside the astronomy project, Macro Grants announced a $100,000 amateur astronomy grant to fund similar work worldwide.
The nonprofit’s review board supports projects such as LIDAR archaeology flights over Central American jungles and digital indexing tools.
“Everyone gets much more excited when it becomes so easy to translate dollars into discoveries,” Depue told BigGo Finance.
Frequently Asked Questions
What specific objects were discovered in the public archives?
Depue said that the project found asteroid and dwarf-planet candidates, but the report did not name specific objects, provide a confirmed count, or describe a verification process.
How do applicants apply for Macro Grants?
Applicants are instructed to text the team directly, with decisions typically returned within two to three days for grants ranging from $500 to $25,000.
Why did NASA not find these objects first?
NASA and other institutions focus their staffing and telescope time on higher-priority targets rather than systematically combing through archival data for small solar-system objects.
Related reading