Claude Found New Dwarf Planets in Public NASA Data | Will Depue artwork

Claude Found New Dwarf Planets in Public NASA Data | Will Depue

MTS

November 14, 2026

Will Depue joins Theo Jaffee to discuss using Claude and Codex to comb through public telescope data for new asteroids and dwarf planets, which inspired his new nonprofit Macro Grants.

0:00

Will Depue

Me and some friends, Luke Verriter and Josh Wolk, have been, for the last few weeks, just playing around with Claude and Codex and trying to look for new objects in the solar system. It's been super, super fun, and we've been just completely shocked with the amount of things you can actually discover with some real effort and digging in with these agents.

It's been super fun. The project started because we were looking for new planets, specifically like the Planet 9, which would be the ninth planet in the solar system. There's been theories about it existing for a long time, and it ended up into finding all these asteroids and finding these dwarf planets and a bunch of other things.

0:39

Theo Jaffee

We are back. We are live with Will Depue, who needs no introduction. Welcome.

0:44

Will Depue

How's it going? I'm happy to be here.

0:48

Theo Jaffee

Tell us about what you announced today. You announced a $100,000 grant for amateur astronomy research.

0:55

Will Depue

Yeah, no, me and some friends, Luke Faradar and Josh Wolk, have been for the last few weeks just playing around with Claude and Codex and trying to look for new objects in the solar system. It's been super, super fun, and we've been just completely shocked with the amount of things you can actually discover with some real effort and digging in with these agents. It's been super fun. The project started because we were looking for new planets, specifically like Planet 9, which would be the ninth planet in the solar system. There's been theories about it existing for a long time, and it ended up into finding all these asteroids and finding these dwarf planets and a bunch of other things. I thought it was a great way to help sponsor other people to go do this.

I'm helping fund a grant and raising money in a non-profit to help fund this type of work around the world.

1:45

Theo Jaffee

How exactly do you find stuff with Claude?

1:49

Will Depue

Yeah, I mean, so pretty much all these telescopes that are on the world have been funded by the US government, European government, Japanese government, et cetera.

These academic public partnerships, they take a lot of them produce massive amounts of data, taking photos of the sky every single night whenever it's clear, whether it's in space or not. And they publish all this data online that's publicly accessible. Most of these surveys, people will be looking for a certain object, trying to make a certain detection and point their telescope at a certain part of the sky. And even if they do accomplish their goal, they usually don't mind that data for a bunch of other analyses, right? Like people are doing targeted data collection, looking to solve certain, you know, doing astronomical research. And so there's just tons of this data out there, and a lot of it hasn't been totally, you know, deeply finally come through because people don't have the time to kind of look for every possible object that appears in any of these data sets. And so, yeah, this is a lot to find, especially now today, where the cost of doing a ton of hardcore software engineering data analysis is just so low and, you know, compute is so abundant.

2:51

Theo Jaffee

Yeah. Why didn't NASA find it? You know, they have the same Claude access that you do.

2:55

Will Depue

But one, I just think that, like, you know, everyone's talent constrained. And I think the great thing here is that you can leverage your tokens and your ability to go kind of look for stuff that you're interested in. I mean, there really is just so much to find. And so I think, you know, a lot of, you know, astronomical research is focused on finding, you know, more important things and other random asteroids. I mean, the dwarf plan discovery is genuinely exciting, but you know, a lot of our stuff is just looking for, you know, the long tail of objects in our solar system that are fun to find.

3:21

Theo Jaffee

I don't know, this is kind of breaking my intuitions in that, like, you know, when it comes to math, for example, a lot of the novel LLM math research has come from the labs themselves.

3:33

Will Depue

Yeah, I mean, I don't know, but there's so much math to do. Like, OpenAI only published the, like, you know, top 400 most important results they got. Like, I'm sure there's a massive number of, you know, unimportant results they didn't publish or unimportant results that people independently are discovering.

And so maybe the largest stuff will continue to be discovered by these institutions, but, you know, you can find some yourself.

Read the other 16 minutes of this episode here.

$15 gets you this episode and 99 more. Credits never expire. After checkout, one link brings you back here to read it.

Not what you expected? Full refund within 14 days.

Thousands of transcripts fetched by people building searchable podcast archives

Fetch transcripts with code

The same credits work from a script or an AI agent. The demo key returns a sample episode in full, no card needed:

request
curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.

Using your own key:

request
curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000801951927