Venture Everywhere Podcast: Nathaniel Manning with Scott Hartley
Nathaniel Manning, co-founder and CEO of LGND AI chats with Scott Hartley, Managing Partner at Everywhere Ventures, on episode 87: What on Earth? Ask LGND AI.
In episode 87 of Venture Everywhere, Scott Hartley, Co-founder and Managing Partner of Everywhere Ventures, sits down with Nathaniel Manning, Co-founder and CEO of LGND — a company making Earth data intuitive and actionable through transformer-based geographic embeddings that enable teams to create, adapt, and scale geospatial datasets across time and geography. Nat shares LGND’s approach to unlocking the untapped value of hundreds of petabytes of satellite imagery, making Earth data both human- and machine-readable. Nat also discussed how AI-powered Earth observation is opening massive market opportunities while shaping mission-driven teams built on clarity and purpose.
In this episode, you will hear:
LGND's use of transformer models to compress satellite imagery into geo embeddings.
The shift toward AI-driven building blocks as the foundation of geospatial analysis.
Building intuitive interfaces that let analysts fine-tune models through yes/no feedback.
Integrating Earth observation with weather and infrastructure data.
Exploring on-satellite processing to send embeddings instead of raw imagery.
If you liked this episode, please give us a rating wherever you found us. To learn more about our work, visit Everywhere.vc and subscribe to our Founders Everywhere Substack. You can also follow us on YouTube, LinkedIn and Twitter for regular updates and news.
TRANSCRIPT
00:00:00 VO: Everywhere Podcast Network.
00:00:14 Jenny Fielding: Hi, and welcome to the Everywhere Podcast. We're a global community of founders and operators who've come together to support the next generation of builders. So the premise of the podcast is just that, founders interviewing other founders about the trials and tribulations of building a company. I hope you enjoy the episode.
00:00:33 Scott: Hi everybody, I'm Scott Hartley, co-founder and Managing Partner of Everywhere Adventures. I'm super excited to be here today with really long time friend of mine, Nat Manning.
00:00:43 Scott: The thread through your whole career has all been in and around geodata, all the way back to your role as CEO of Ushahidi, which was a crisis mapping application in East Africa, to running open data under the Presidential Innovation Fellows Program at the White House for USAID and being the Chief Data Officer at USAID, to the founding of Kettle Reinsurance, using geospatial datasets across time and geography to better manage risk around catastrophic events like wildfires, given what's been happening over the last few years.
00:01:16 Scott: To your current role as the CEO and co-founder with our friend, Dan Hammer of a company called LGND. LGND is really building Google Earth for the AI generation perplexity across all things geodata sets. Welcome to the podcast. Super excited to have you with us today.
00:01:32 Nathaniel: Thanks for having me, Scott. Good to be here with you.
00:01:34 Scott: Tell us a little bit more about LGND. I know that Dan Hammer, your co-founder who was running all things APIs and data extensibility for NASA had a deep geospatial background as you. You guys were working on this open data project called Clay. Walk us through the genesis of LGND and what led you guys to this idea.
00:01:53 Nathaniel: I was working, as you said in the last iteration, built this company, Kettle. The thesis was that climate change is going to break property insurance markets. And two, you could use machine learning or AI to be able to run it on satellite imagery or weather data and better predict risk, ones that were being exacerbated by climate change, like wildfires and hurricanes.
00:02:12 Nathaniel: Both those thesis ended up proven really true and it worked really well. And in that work, we probably spent millions of dollars training CNNs (Convolutional Neural Networks) on questions like could we find all the fire breaks in California and then many other ones? And then that created an ensemble model. And each one of these CNNs, you’d feed it tens of thousands of images to train and over one query like that. And it worked really well.
00:02:34 Nathaniel: But then this little thing happened. ChatGPT came out, 2022, and we were in it, and that company has sold insurance and it went well and eventually passed the baton because the next phase is really about scaling an insurance company. As a technologist it felt it needed a real insurance head is what made the most sense for the next phase.
00:02:51 Nathaniel: I was sitting back just thinking about this technology and what had changed since ChatGPT came out. At the same time then, I went and caught up with my old friend, Dan. Bruno's our third co-founder. Dan and Bruno had been building this fully open source model called Clay. They had the same thing.
00:03:05 Nathaniel: Right after ChatGPT came out, they said, hey, could you apply this transformer model architecture, which is the intention is all you need. It's the innovation behind everything that's become LLMs. An LLM stands for Large Language Model. Instead of a large language model trained on language, could we train it on a large earth observation model?
00:03:27 Nathaniel: Clay is a large earth observation model, fully open source and open weights, all done under a nonprofit for the purpose of climate and environment with donated compute. It's a GitHub project. It's out there. They'd been working on that.
00:03:40 Nathaniel: Then we were all catching up and realizing that those two experiences were the two threads of the DNA helix that became LGND because people started saying, hey, Clay's cool. Can we put this to work? But most people don't know how or want to go grab a model and do everything for it themselves.
00:03:54 Nathaniel: And then in our space over here in Earth Observation and Geo, there's no open AI enterprise tool. There's no LangChain yet. There's no infrastructure to put all of this to work. It's like 2020 in LLM land.
00:04:06 Nathaniel: It became very clear to us that there was a need to build this bridge between these emerging models being trained on Earth observation data, not language, and being able to put them to work easily. That was the formation of the company. The real goal of what we're trying to do is being able to create a tool that lets you query the Earth over space and time.
00:04:25 Scott: Querying the earth over space and time. It reminds me of when I had the chance at Fika Ventures offsite a couple of years ago to meet one of the co-founders of SpaceX. I asked him, what problem are you solving? And he said in the simplest terms, gravity. And I thought that is the most crystal clear definition I've ever heard.
00:04:44 Nathaniel: Right.
00:04:44 Scott: Talk about querying the Earth's data over space and time has a Google-esque clarity to it or a SpaceX-esque clarity to it, which I love and it's a huge vision. I remember we had lunch at Tartine in San Francisco. And you said to me, I knew that your background had been in and around crisis mapping and a lot of emerging markets, and we shared that passion having spent a lot of time in East Africa.
00:05:06 Scott: You said to me, "I'm thinking about going into insurance." It was a mic drop. I didn't quite understand the prescience that you had or the foresight that you had thinking about all of the things that you learned in and around mapping and in and around these emerging market data sets, in and around USAID and open data and thinking ahead to these applications for risk modeling and for insurance.
00:05:29 Scott: You know Steve Jobs' speech about staying hungry, staying foolish, he also said in that speech that the dots don't make sense going forward, they only make sense going backwards.
00:05:38 Scott: As you guys were building Kettle and you were looking at how do we get more granular risk assessment to better model and better underwrite risk in a wildfire edge case, it should be way more granular than a zip code.
00:05:51 Scott: It should be based on which side of the hillside are you on? What are the wind patterns? Where are the fire breaks? All these earth level data sets, it makes sense to me as you explained that, that you guys really discovered this market for LGND in some ways out of the problems you encountered with data modeling for Kettle, is that right?
00:06:11 Nathaniel: Yeah, I think it certainly opened the door to it. I was in a very applied space for the area, highly verticalized like Kettle. Kettle sells insurance policies. It doesn't sell technology, but we've got all of our own technology ground up of use of risk.
00:06:26 Nathaniel: What LGND allows for and what is usually done with these images of the earth. When I say earth observation, I mean satellite imagery for the most part but it also could be low earth flying planes. It could be drone imagery, but it's basically the top down view. Just think about it that way.
00:06:42 Nathaniel: Most of the time when folks are using this observation, what you're doing is you're taking these complex pictures and trying to translate them into information. And information is often semantic, right? That's what we understand as humans or in a grid type format. But us humans are really good at doing that.
00:07:00 Nathaniel: We're good at looking at a picture and saying, okay, in this picture, there's a bunch of properties, there's a big old eight lane highway and then there's a forest on the other side and saying, oh, that eight lane highway, that acts as a fire break between that thing that's very burnable and those homes that have property.
00:07:17 Nathaniel: So that's what a traditional underwriter would do. That's what an analyst would do. That's stage one of how you create knowledge and information out of pictures. Cause ultimately these are pictures. And phase two was what I talked about earlier, where you'd have a very expensive couple hundred thousand dollar project.
00:07:35 Nathaniel: And this is where most of Earth observation as an industry has been for the last bit of time, where you would try to glean knowledge through doing a very specific, trained CNN on one query. That would take a couple of data scientists and an MLOps person and an infrastructure to be able to do that.
00:07:53 Nathaniel: People talked a lot about counting cars and Walmarts and all sorts of other examples. This has been done a ton in agriculture. It's done and used in carbon accounting. It's done and used in government and defense work, but it's still the same thing. You're ultimately trying to create a data set out of these pictures.
00:08:11 Nathaniel: And then what has just happened in applying all the technology that's behind LLMs and into this data set is the ability to do that orders of magnitude faster because you have a pre-trained model and that's what this pre-trained model is doing.
00:08:28 Nathaniel: It's able to do the same thing of saying that eight lane highway is what we think of as a firebreak and label it and organize it and create that insight, build a data set in like milliseconds from a query by saying, can you create me a map of all the firebreaks in California? That doesn't exist.
00:08:48 Nathaniel: You can't ask ChatGPT today, can you create me a map of all the firebreaks in California? But the answer to that question does exist. It exists in pixels, in open data that has been available back to what we used to do. You know, that dots looking backwards only makes sense. That's the part that always stuck with me from Jobs' speech.
00:09:07 Nathaniel: Dan, Bruno and I all met back in 2012-14 timeframe. Bruno was the chief scientist at Mapbox. Dan and I were both working in opening up government data and making it more available.
00:09:19 Nathaniel: That's the whole point is because this data that NASA or the European Space Agency has been collecting and making available and Dan helped build some of those APIs and put them out there, listing out the fire breaks one. The answer to that question exists, but because ChatGPT or any LLM is trained on language, the answer to that question doesn't exist in language today. It exists in the pixels. That's what we're able to bring to light.
00:09:41 Nathaniel: And we think that there's a lot of value in that that's been really untapped because it has been such a hard stack to work with. It's a complicated set of data. I mean, to put in perspective, I think this amount of data doesn't necessarily equal value, but it does equal complexity and maybe some bit of potential value.
00:10:00 Nathaniel: All the language in the world, like what all these LLMs are trained on, is sub one petabyte. I think DALL-E and all the clip image generations are three to five petabytes of data of pictures that have been out there that people train on.
00:10:13 Nathaniel: And we've collected something like 100 to 200 petabytes of Earth imagery, but only like 200 times more volume, which both makes it extremely heavy, difficult stuff to work with versus semantics. I think that's both why there's so much potential.
00:10:29 Scott: Just the fact over the last decade plus in thinking of what's been transpiring with SpaceX and in bringing the cost per kg down asymptotically to zero. If you think of that, it's almost building a railroad to space. It's building the railroad West. Instead of go West young man, it's go up.
00:10:48 Scott: And we're starting to put more and more things into orbit. Google Earth was created out of Keyhole. There were a number of planet labs doing early optical imagery. I saw recently Spire Global, which runs a lot of different forms of geodata around ADS-B, which is flight tracking, AIS, which is maritime tracking, GPSRO, which is a lot of weather data.
00:11:10 Scott: Starting to integrate data sets, starting to think about ADS-B, so flight tracking data interfaced with FAA calls and calls about turbulence and mapping and creating context on top of weather data and on top of flight data to figure out where are actual areas of the earth that are high probability of turbulence.
00:11:31 Scott: And that's the point of layering context on top of pixels, layering context on top of raw data. In our portfolio, we have a company called Satim out of Poland, which is doing this specifically around asset tagging, mostly around military assets, but maritime and military to be able to take a synthetic aperture radar image, SAR image that can be taken through cloud cover and through darkness.
00:11:57 Scott: In addition to optical imagery, which can only happen in that 25% of the time when it's not cloudy and it's daytime, but taking that set of data and then being able to tag it with maritime data, tag it with ship names, tag it with information like that.
00:12:09 Scott: There are a lot of these vertical specific companies being built around data tagging within specific domains. Are you guys partnering with some of those players out there?
00:12:19 Scott: Do you see all these raw data feeds as piping into LGND or how do you think of this, as this ecosystem evolves, starting with the price per kg going down to zero, more things going into space, more data being collected in LEO, Low Earth Orbit.
00:12:34 Scott: How do you think of not just the data sets that exist – this is the tip of the iceberg as this set of data explodes even further as those costs go down. But how do you think about that from a market standpoint?
00:12:45 Nathaniel: Our thesis is this, which is that all large data sets are going to have transformer model architecture applied to them. We've seen that happen in language. We know the winner's there. We're seeing it happen in self-driving. We know the winner's there.
00:12:58 Nathaniel: We've seen it happen in image generation and more recently, like audio, the live kits and chatterboxes out there. The largest actual data set of all of those from, as I just said, there isn't a synonymous name with that for AI today. And that's what we aim to be.
00:13:15 Nathaniel: So what happens when you apply the transform model architecture to data is the output is an embedding. And an embedding is in short, it's a vector of a string of numbers. It's essentially compression. And I talk about it as a metaphor of like, it's like a fingerprint. A fingerprint can tell you who somebody is.
00:13:31 Nathaniel: It has super low data, but that fingerprint is to one person and all of the information you might have about that person. It's a giant compression of identity in one simple thing. And that's what an embedding does for language. There's more to it. It talks about how it works in context of other things.
00:13:48 Nathaniel: But for us, the thesis is that for 20 to 25 years, keywords were the primary data source for what I think of as a first order data object for organizing and making sense of language. About the same amount of time, MAC tiles or raster data was the first order data object for making sense of Earth observation.
00:14:10 Nathaniel: In the last 30 months, that has changed from keywords to language embeddings. That is now the first order data object for organizing semantic knowledge. We just think that the same thing is going to happen to the Earth observation, to the geospace. We're going to transfer from pure pictures and MAC tiles to geo embeddings for all the same reasons.
00:14:30 Nathaniel: We're trying to help usher that in. In the article, TechCrunch put out about us, talked about us as the standard oil. It's ironic because we all come from climate backgrounds and environmental backgrounds, but what was specifically meant there is what standard oil did really well was refine crude outputs, you know, raw ore and refine it into something trusted and usable and repeatable.
00:14:51 Nathaniel: That's essentially what LGND is aiming to do is compress and refine all of this data that's out there. We love all of the satellite companies, public and private or drone companies or who are creating this content and imagery content material.
00:15:06 Nathaniel: And we need to partner with them and be helpful because hopefully we can help build the on-ramps to using them that much more easily and make it speak the language of AI. One of the taglines I like to talk about for us is we're trying to use AI to make Earth understandable. We're also trying to make Earth understandable to AI. And that's what we do.
00:15:26 Scott: It's really fascinating to think about making things both human readable, but also machine readable. It seems like there have been a few attempts over the years to compress, as you say, observation data around earth or location pinning to specific things.
00:15:40 Scott: One company that comes to mind from many years back was trying to pin locations on the map with three specific words, but that was a somewhat human interface to get to hyper-specific mapping locations. But there've been a number of attempts to consolidate over the years.
00:15:56 Scott: I love that metaphor of a fingerprint and in many ways the world always accordions in and out and there's expansion and compression. And it seems like we've gone through a boom of expansion where this level of data acquisition, this level has gone so far to the point of you seeing, you know, 200 petabytes worth of imagery data.
00:16:14 Scott: And there has to be some means of consolidation to make it accessible, extensible, able to build on top of. And I think that's you guys' backgrounds and Dan's in particular of doing this with NASA datasets that becomes really interesting as a new ground zero.
00:16:30 Scott: One following question on this would be the data that's collected from space, the subset that's even downloaded or on Earth is less than what's actually collected in space because of the compression challenges, because of overflight times with ground stations and limitations on bandwidth.
00:16:46 Scott: There's a number of companies being built to expand bandwidth to enable more downloading of data from space. There's also companies like Starcloud in our portfolio that aim to do compression and data storage in space and be able to do compute in very low Earth orbit rather than downloading to Earth.
00:17:03 Scott: As you think about the frontier of where the world goes over these next few years, are you guys thinking about some of those really far-flung ideas like data centers in space, the expansion of more Earth observation data, going even more bananas than it already has over the last decade?
00:17:18 Nathaniel: I would love to talk to that other portfolio company. Yes, we have, and it has come up with a couple of the satellite companies out there just like, yeah, you could run LGND. We're not billing, but you could run it on site, on-prem, on a satellite and send embeddings down, which would be orders of magnitude less heavy.
00:17:37 Nathaniel: Our aim with these models and things is for them to have minuscule data loss, right? Minuscule information loss. That would work tremendously well. It has come up a couple of times. That is absolutely a use case that we haven't talked publicly about yet. So that's pretty fun.
00:17:50 Nathaniel: The other one I was remembering you were saying earlier about weather data and things too. So I talked about our standard oil for these heavy objects into refined embeddings. And that's part of our thesis that embeddings is all going to go in one direction and that we're going to be the shot for building the top down view of the world into embeddings.
00:18:07 Nathaniel: The second is to talk about the weather data and other things is we're also very much experimenting with how does that interact with other data sources? We're not going to try to build our own weather foundation model. People have done incredible jobs.
00:18:18 Nathaniel: Doing that, it's not our competitive advantage. But what's so cool in the world right now, one example would be MCPD servers and things letting you access all this data, like how are these integrating all this API.
00:18:30 Nathaniel: So we are absolutely experimenting with the other side, I would say the front end and being able to integrate other insights as well. And back to the firebreak example, the aim, right, is being able to say, “Hey, help me find all the firebreaks.” And you have an interface that helps you instinctually know that you build that data set.
00:18:47 Nathaniel: But then you could ask it, “okay, weed out anyone that's had rain in the last three months” or “color in blue anything that you think is going to get rain in the next month.” And then with points of interest data, you could say, “call out any of them that are under 10 miles from the fire station.”
00:19:04 Nathaniel: I'm still obviously thinking like a wildfire underwriter, but there's like some of the stuff that it's all out there. I have these incredible conversations with LLMs that can get you a similar experience asking about all this wild stuff that's known in language, quantum physics, or how some of these models work because they're written up.
00:19:23 Nathaniel: But the answer to those three questions, they are not in language in the same way. And that's what gets pretty cool.
00:19:28 Scott: Are there human data taggers in the loop that are helping stitch together these pixels to linguistic applications? Like you talk about rain impacting something or fire break.
00:19:41 Scott: These words have context, have meaning and have an application or a scope that applies to pixels, but that scope can't be known unless that data is first tagged and incorporated into the model.
00:19:54 Scott: As you guys build what is really more perplexity for geodata rather than open AI for geodata, as you mentioned, because it's an open set of ability to query across multiple models and multiple things under the hood rather than one closed ecosystem. But how does data tagging get into this in humans in the loop?
00:20:13 Nathaniel: One of the reasons why all this works now is because you can build off of all of the tagging that's happened in the last 20 years. So the models get trained off of that. And then what our product allows you to do is do that fine tuning for the rest of the tagging at the end.
00:20:30 Nathaniel: Let's stick with the same example for all the lat longs of all the fire breaks in the state. And let's say I want to turn up and you go, hmm, really through a direct interface, being able to click yes and no, fine tuning the data set at the end.
00:20:45 Nathaniel: And it's an easy to do tool versus what currently happens, which is like a pretty highly skilled data scientist and ML engineer doing that fine tuning of the model under that CNN example. But here, an analyst who's trying to find this answer would be able to do it through an easy interface.
00:21:01 Nathaniel: And that fine tuning creates labels. A lot of what we're doing here with the models is creating data.
00:21:08 Nathaniel: If you believe that 99% of the value of what we know about the earth has already been collected, it is basically where's the Starbucks and how to drive there, which is a problem that has been solved. I give all the credit in the world to Google Maps and Keyhole team who helped back us, which is why they're tremendous, who solved that problem.
00:21:30 Nathaniel: And then Mapbox came along as well and did it outside of the Google-verse, who I also think the world of. But that is points of interest in driving and navigation. And if the belief is that is 99% of the dollar value of the physical world, then we're not going to be that successful. We don't believe that.
00:21:46 Nathaniel: We believe that there's a lot of value in knowing a lot more about the blue and green blobs of the world over space and time. The mission is we're engineers and we believe that in that adage, you can't fix what you can't measure. And that's ultimately what we're trying to help fix.
00:22:02 Scott: Shifting gears just in the last couple of minutes of the podcast here. You also run Kindergarten Ventures, which you started, which enables you to see and talk to a number of founders like yourself and invest in around these themes.
00:22:14 Scott: And as a CEO yourself over multiple companies, you talk a lot about leadership and how leadership is setting vision and removing obstacles and those things that sounds so simple, but are so actually hard to do in real life.
00:22:27 Scott: As you evaluate CEOs that you're going to invest in, or as you think about your own journey, any takeaways or any learnings from how you do that well?
00:22:35 Nathaniel: I think there's many ways to run a company. First off, everyone has different approaches and it can work in many different ways. What I've learned is the ability to build a world-class team. I ultimately think this is a team sport and then it's the ability to look at how that team works together and diagnose and fix bottlenecks.
00:22:54 Nathaniel: People want structure and everyone's like, oh, ‘I do not want TPS reports.” No one wants TPS reports, very true. We don't want bureaucracy, but people want structure. They want to know where they're going and how to get there and how they best contribute towards direction. That's a balance that you constantly have to be balancing.
00:23:10 Nathaniel: And then you need to be constantly trying to figure out where the bottleneck is and fix it. It's a constant plugging of holes game and it has to be fun. And that is a question of team efficiency, velocity, talent density, and working as a soccer team on the match.
00:23:25 Nathaniel: We're a soccer team, we're not a track and field team. One person can't win a race and it's a win. And if the team loses, it doesn't work that way. I think CEO's job is to keep that in mind and the operating in that way. And ultimately, I think as an investor, back to that quote, where the dots looking backwards makes sense to the next step. They're obsessed with this thing.
00:23:45 Scott: Amazing. Such a rich set of experiences that you've had through those dots looking backwards. And I'm thankful that many of those dots from The Presidential Innovation Fellows Program at the White House in DC to overlapping in Kenya together, to investing in Kettle with you, to investing in LGND with you. You had four or five touch points with you over the last 15 or 20 years, which has been a true joy in my life.
00:24:07 Nathaniel: Me too, Scott. It's been really fun. We've gotten to a lot of fun stuff over the years. Scott is amazing and has been an incredible resource and friend and advisor, all sorts of ups and downs.
00:24:16 Scott: Looking forward to the next decade plus. Finally, where can listeners find you online?
00:24:21 Nathaniel: LGND is LGND.io. For me, it's Nathaniel Manning on LinkedIn or Nat, natpmanning on X. Twitter, I still call it Twitter. That's where I'm most of the time, those two places.
00:24:35 Scott: Any book or podcast that you're currently listening to or recommend for us?
00:24:39 Nathaniel: On the podcast train, Kindergarten Ventures is me and David Rosenthal from Acquired, biggest fan here there is. I thought the Indian Premier League was just tremendous. It was like such a story I didn't know, such a hero's journey. So that was fascinating. I loved that.
00:24:55 Nathaniel: Book wise, so many good books. Honestly, as a leader, there's a psychology practice called "parts work" or Internal Family Systems by Schwartz. And I was reading one of his books recently. I thought it was really revelatory. And I think kind of important as a leader as well.
00:25:12 Nathaniel: You realize it's often in our language, like this part of me is really upset about this, but this part is affected by this and you go, oh, it's good to know and be able to recognize that. I think it's a helpful self-improvement and self-awareness as we all try to constantly be improving over time.
00:25:25 Scott: Thank you for sharing that. We'll include the links in the transcript of the podcast. Nat, thank you so much for joining us. Thanks for your time. Thanks for everything you're doing with LGND and we're super stoked to be small ambassadors and a little part of the journey with you.
00:25:38 Nathaniel: Of course. One of the first people I called. Thanks so much.
00:25:41 Scott: Thanks, Nat.
00:25:42 Scott: Thanks for joining us and hope you enjoyed today's episode. For those of you listening, you might also be interested to learn more about Everywhere, where a first-check pre-seed fund that does exactly that invests everywhere. We're community of 500 founders and operators, and we've invested in over 250 companies around the globe. Find us at our website, Everywhere.vc, on LinkedIn, and through our regular founder spotlights on Substack. Be sure to subscribe, and we'll catch you on the next episode.