reading
21 min read

The abundant, cheap energy sitting at the bottom of the sea

A conversation between Graham Littlehale and Andrew Redd on how Endurance Energy is tapping subsea geothermal energy

Graham Littlehale

When Andrew Redd saw capital start flowing into the AI buildout, he knew the end result would be more natural gas drilling if no one came up with a better renewable solution. His company, Endurance Energy (opens in new tab), has completely rethought how – and where – we get energy. And that’s required reimagining the sea floor.

On land, geothermal energy has struggled: most sites lack the requisite heat, water, and permeability. But over 80% of the world’s volcanoes are underwater. They sit under thin, porous crust surrounded by ocean. Endurance Energy believes that subsea geothermal can be the cheapest electricity ever produced, and faster to deploy than natural gas.

I sat down with Andrew to discuss how Endurance is tapping the sea floor, the results of the team's first two missions, and their plans to use Tonga’s faulty electric grid as their first large proof of concept. They also spoke about Andrew’s background at SpaceX, and how he approaches hiring and developing a team of engineers tackling one of the world’s most ambitious problems.


Transcript:

Andrew Redd, CEO and Co-founder, Endurance Energy

Graham Littlehale, Partner, Felicis

Graham Littlehale: Andrew Redd, welcome to Felicis.

Andrew Redd: Thanks, Graham. Thanks for having me.

Graham Littlehale: Of course, of course. For those of you who don't know me, I'm Graham Littlehale. I work on deep tech investing here at Felicis. I invested in Andrew's company, Endurance, back at the pre-seed almost two years ago today. So Endurance will be new to a lot of our viewers. Can you tell us what exactly is Endurance Energy?

Andrew Redd: Yeah, Endurance develops subsea geothermal power. We believe that subsea geothermal electricity can be the cheapest form of electricity ever. So we're tapping into heat in Earth's crust at mid-ocean ridges. So what we're doing is drilling shallow wells into those mid-ocean ridges, extracting hot water, converting the hot water into electricity, and then bringing that electricity back to shore to power island nations, industrial sites, data centers, coastal grids. Really what we're able to provide is clean baseload electricity. Baseload means always on 24/7 power as opposed to an intermittent resource.

Graham Littlehale: So you're no stranger to difficult problems throughout your life. I think when you were first introduced to me, I was told Andrew was a rower at Princeton, so it means he likes to eat glass. And then you went from there to Neuralink and then SpaceX.

Andrew Redd: Right.

Graham Littlehale: And then you decided to found Endurance. Talk to us a little bit just through your background, some of the problems you've worked through on your career - during your career, and then why you decided that offshore geothermal was the right next thing.

Andrew Redd: Yeah, I think I do definitely love to chew glass. That's absolutely true. I think that I've kind of been searching for the hardest possible positions to put myself in, regardless of in what form. And so going from college to Neuralink - a great way to cut your teeth after college as a 40-person Elon startup. Definitely got to experience that. And then moving to SpaceX was another big opportunity of going and seeing how a group of people can come together to create something truly world-changing and monumental. So that was kind of a wonderful thing to witness of how to go from zero to one and how to create something bigger than any individual really could imagine. And you can only do that through extensive teamwork and having kind of boundless ambition.

Graham Littlehale: And then how did you get the idea for offshore geothermal? What was the inception point? And then how did you decide this was what you wanted to do next?

Andrew Redd: Yeah, I think the common thread throughout my career has been working on safety-critical systems in extreme environments. So in college, I worked on spacesuits and satellites. And then at Neuralink, the inside of someone's brain is definitely an extreme environment. And then at SpaceX, working on Starship and working on Dragon kind of continued to hit that theme. And the element there that is most essential is engineering rigor and intellectual honesty.

I was seeing the data center build out and all of the capital going into AI infrastructure and saw pretty clearly that power availability was the limiter there. It didn't take much to show that unchecked and without an alternative solution, natural gas generation was going to support basically all of that future build out. And so I really saw it as my mission to provide an alternative that could be cheaper and faster to deploy than natural gas. And I was able to back up into offshore geothermal with that general framework.

So I really needed this energy source to be renewable. That was important for founding a future company for me. I want to do something that is good and positive for the world. I don't want any cognitive dissonance going to work every single day. You have to have baseload electricity. That's kind of table stakes for any data center development, industrial site or grid authority. You need that firming power source. You need it to be actually deployable quickly. Availability is not just about can you produce power, it's can you produce power and get it out into the world quickly. So availability means low permitting overhead. It means power density. It means transportability. It means this shouldn't be a construction project. This should be happening in a factory.

And then the fourth is scalability. If we really are talking about re-industrializing the U.S., building out hyperscale data centers, supporting future electrification of our grid, of all electric cars, and bringing back industry, then you need to be talking in the tens of gigawatts scale. So those were kind of the four elements that I had to put together. Renewable, baseload, actually deployable, and then scalable. Endurance's approach is to go where the resource is already scalable. The vast majority of volcanoes are offshore. It's something like 80% or 90% of all volcanoes are offshore. And so if you're able to extract geothermal power in the subsea, then now it has all those factors that we already discussed, but now it's scalable. And with scale, you can reduce costs because you're able to replicate over and over and over again.

Graham Littlehale: What makes a site ideal for Endurance to generate offshore geothermal electricity?

Andrew Redd: You need permeability, you need heat, and you need water. On land, you can find maybe some hydrothermal reservoir that has all those three things, but we've kind of tapped all those resources out on earth. And so now technologies are going to creating permeability and bringing their own water, right, or going deeper for the heat. Offshore, you have permeability because you have really, really young crust from the mantle freezing and, you know, crust less than two million years old. It's that kind of pillow lava, that permeable basalt that you find in Hawaii.

And then you have your, you have infinite water, right, from just the ocean that's constantly under pressure being pushed into the permeable rock, and then you have your heat, right? Oceanic crust is an order of magnitude thinner than continental crust. So you've got the holy grail of geothermal. You've got three elements, and not only are they there, they're there in spades, and they're there in a consistent way over thousands of kilometers.

Graham Littlehale: You know, maybe just walk us through actually, since founding the business two years ago, what deployments you've done, and then talk about the most recent one, which was super exciting. It was your Adélie deployment, where you went down and actually generated electricity on the seafloor.

Andrew Redd: Yeah. So Endurance was founded in January of 2025. And since the first kind of six months of the business was really building out the core team, and also validating the technoeconomics of the idea, right? The team I'm building are such capable engineers that they can, they truly have the capability to change the world, and they should be working on something that's extremely important. So validating that this is the right thing to go after was really important to me in order to scale the business. Between August of 2025 and now, we have been just, you know, once the thesis was validated, we've been iterating and deploying at an extremely fast rate. So in that sort of, you know, I guess 14-month period, we've done five deep-sea deployments to different volcanic systems worldwide.

And then most recently, we went back to the Juan de Fuca Ridge, one mile deep to the Axial Seamount, which is an active volcano and caldera on the seafloor, and did our Adélie deployment, which is our first end-to-end test of subsea geothermal, all the way from drilling, really resource access and production to power generation, and then, you know, transmission back to shore. In this case, we transmitted data, but we were communicating with the system from all the way from Seattle.

Graham Littlehale: So I feel very fortunate that I was able to watch a lot of this deployment live streamed, which for me, you know, I never worked at SpaceX, but it was like, I felt like a kid in a candy store, like if I could like rewind 20 years and go watch early SpaceX launches as they were happening and watching the live stream of you guys deploying onto the seafloor was absolutely unbelievable in like high-fidelity video too. I guess just talk a little bit about what the Adélie deployment took to make happen and like what you did when you were down there, because, you know, the headline is that you generated electricity, but there's so much that the team poured into that. And it was a multi-day mission. It was not just, you know, you go in for three hours and you're done. So yeah, just give us the play-by-play on what really happened there.

Andrew Redd: Yeah. So it was a three-week-long deployment. And that whole time we were full pedal down operating at full tilt, either solving problems or doing operations on deck or in the deep sea. The majority of the offshore deployment ops with the drill and the generator was run by Endurance employees. So we were ensuring that all of that was done safely and effectively without damaging any humans or any hardware. That adds a whole nother level of complexity to traditional R&D projects.

So the first leg of that mission was deploying our subsea lander drill, which is a seven-ton system that's capable to 80 meters of drill depth. It's a two-inch borehole. So it's incredibly small in terms of impact to the seafloor. So, you know, in those first week and a half that we're deploying the drill, we would put the system in the water, detect the ground fault, bring it back up, put the system in the water, get it to the seafloor, start drilling, find some issues with the mechanicals, bring it back up, fix those. And so it was really an exercise in continuing to iterate and continuing to move the ball forward just to get to that next level of technological development. And that's kind of how it feels a lot of the time with doing things that no one's ever done before and that are really quite brand new is you're just pushing for every, you know, additional bit of mileage.

Once we got that system to the seafloor and started drilling wells, then it became, okay, how do we drill a well that can be utilized by a generator to actually produce power? How do we get the most amount of resource from this well while not, you know, getting the drill stuck on the seafloor or, you know, ending up with nothing? On our second well, we got to 16 meters deep and saw that we were definitely producing fluid from it. And so we decided, hey, let's stop this and let's case and complete this system. So, about one-fifth of the total capability in terms of depth of the drill. We're able to put in casing, put in a ball valve on top.

We're able to produce, you know, the world's first cased and completed subsea geothermal well, had 261-degree fluid coming out of it. It's about 400 kilowatts of total thermal power we were able to extract from that system. And then actually brought the boat back to shore to load up a new set of crew, including myself as well as the generator. So then on the second leg went back to the Axial Seamount, went back to our well and deployed the generator.

So we're able to get Adélie onto the seafloor and connect it to the geothermal well we had drilled. It ended up being too restrictive for us to produce power with the geothermal well. So we actually moved it to a nearby hydrothermal vent in the area. And so that was a game-time decision where we're actively changing our mission parameters, our mission plans to give us the best chance of success and to enable us to learn the most.

Graham Littlehale: And you were able to generate electricity for a short period of time.

Andrew Redd: Yeah, correct.

Graham Littlehale: It was - how long was it? 10 seconds, something like that?

Andrew Redd: 8.7 seconds.

Graham Littlehale: Okay. And what do you - what were the biggest learnings you took away from this experience?

Andrew Redd: Yeah.

Graham Littlehale: Like the top couple, top learnings.

Andrew Redd: I think the - so, a couple of things. One, it's validation of all the design decisions that went into the generator were successful. So there's a bunch of elements that go into it, but the primary kind of the core structure of the thing is a large fluid loop. And so this is the heat exchangers, the turbomachinery, the pump, and it's all sealed against seawater. So we have a refrigerant that's actually below one atmosphere. So it's less pressure than here on land. And that has to seal against the seawater the entire time. So it validated really all of our construction methodologies for that primary system.

The other element is the electronics, both the power electronics and the command and control, so the brains and the brawn. Proving that all of those systems that were mostly custom boards could survive at pressure was incredibly useful for us. Other thing we learned is a lot about just the offshore operation. So we are able to deploy and recover multi-ton systems, multi-ton generators and drills and, you know, single digit number of hours. That's hugely useful for when we plan future offshore operations, how we stagger things and how we put these operations together and how fast we can move.

Graham Littlehale: What were some things that maybe surprised to the upside with offshore geothermal on this first like major deployment?

Andrew Redd: I was always a little bit skeptical of the cold-side resource quality. Like it's very nice to talk about the infinite cold sink of seawater. Something that we've learned over our operations is the coldness of the depths is really, really dominating and the effectiveness of our cold-side heat exchangers is as good as it's, you know, sort of chalked up to be. And that's something that we were able to validate over and over and over again.

I think another element that's been quite interesting is accessing the resource at really shallow depth. That's probably the most critical one. The second well we drilled was only 16 meters deep. That's like comically shallow for either offshore oil and gas or terrestrial geothermal. But we had extremely high-temperature fluid coming out. Really what that speaks to is on land you have to drill X number of kilometers to get to a level of pressure that can sustain high-temperature fluid. But for us, all of that pressure comes from the water column above us. And so you can get that really, really high-temperature fluid through very little drilling. And that's something that's hugely validating for the thesis at large.

Graham Littlehale: So I know Adélie was a great mission and you learned a ton, but you're not stopping. You're like immediately going to the next one and you've already started building the system for that. So can you talk about what's next and the deployment that you're planning in Tonga?

Andrew Redd: Yeah. So our Tonga mission is really, we did an end-to-end demonstration of the technology of offshore geothermal, kind of directly addressing maybe the science risk of the project. And now we want to address the much more of the real-world risk. We want to be able to deploy to a grid to actually provide power, provide value to people, reduce energy costs for a population, and show that not only is this feasible, but we can do it and we can do it very quickly.

So Tonga as an island has unbelievably high electricity costs. And this is because it's primarily imported diesel. So they end up paying four times what we pay here in the U.S. for electricity and it's not reliable and it's not domestically produced. And so there's this incredible incentive to create a new form of domestic generation, but there's not really a good solution there. So what we're providing is domestic and baseload resource and we're reducing their energy costs by hopefully a significant amount.

There are some seamounts about 50 kilometers west of Tonga that we've identified some drilling locations. And we'll be deploying multi-megawatt generators to drilled wells out there. And we'll start development of this early next year. We'll run a power cable from that seamount to the west coast of the island where we'll build a land substation, so a little under two acres, and then connect to the Tongan power grid. So we've over the past few months been working directly with the Tongan Power Authority to ensure that the electricity that we're providing them is high enough quality that it can actually serve customers. And that's just a whole other host of challenges and things we have to learn in order to make this a large-scale commercially viable technology.

Graham Littlehale: Cool. So you talked a little bit about how you're bringing together a lot of aerospace engineers to a more traditional problem like oil- and gas-related technology. And it's a bit unconventional to do. You know, when you first started the business, you're a solo founder, you've now grown it to over 40 people. Talk a little bit about the team you've brought around you and what it takes to succeed at Endurance.

Andrew Redd: Yeah, I think something I learned from both Neuralink and SpaceX is that personnel and hiring can be either a positive or a negative feedback loop. If you get A players, they all want to hire A-plus players. And that's something that was really important for me to tap into extremely early on. And so really the core and like the nucleus of the Endurance team that started early 2025 are primarily extremely good SpaceX engineers across multiple sectors and multiple subject matters.

And what ended up happening was the quality of that team and what we were able to do with so little created a reputation in the Seattle area. And really, we were able to use the network effect to continue to accumulate extremely high-quality people. And it's not just technical capability. It's, are these people lovely to work with? And that's really a critical element of the Endurance team is like, low-ego, lovely people that are also extremely good at what they do. And if you put enough those people into a room, then it is the sort of best possible environment for - that an engineer would want to enter, right, and want to spend many years of their life.

So that's how I thought about this problem is how do I create an environment where I want to spend multiple decades there and like I will spend multiple decades on this project. So I kind of need that to be good. And it's a combination of what's the location? What type of ownership structure do people have? How much responsibility do they get? And then really, really critically, who do they work with? And what do they work on?

Graham Littlehale: Yeah, you guys have an incredible office right on South Lake Union, literally at the dock. Maybe if some prospective employees or awesome engineers are watching this, who are you looking for? What types of people are you trying to hire?

Andrew Redd: Yeah, we're looking for extremely high-agency individuals who love to work hard and are excited to put their skills to work on a world-changing technology. I think the people that we like to hire are the ones who can approach the problem from every layer of abstraction in a competent and correct way. And so I think if someone is like, the people that I want to tap for the company are those that have consistently over the course of their career sought out the biggest challenges and then were successful at either achieving them or failing in such a way where they learn.

Graham Littlehale: So how do you build a culture of continuous learning when the stakes are high and the infrastructure is really expensive?

Andrew Redd: Yeah, I think there are two major portions to this. One is on the personnel side, and that's, what culture are you creating at the company? And to us, it's incredibly important that every individual is low-ego and very much focused on the product development and not on finger pointing, whose failure caused what, who made what decision. That's really huge. And just continuously reiterating that and showing that failure is okay, and you move on and you figure out what you can learn from it and you continue.

I think the other element comes down to the architecture of the product in a lot of ways of yes, the infrastructure is very large and very expensive, but if you can break it down into individual components and push all of them in parallel, then you can learn from everywhere, even if something falls completely flat and something else develops in a big way. So I would say every large infrastructure project should be broken down into individual components, which are sort of tested in their own right. And that enables each failure to feel a little bit smaller, but you're still learning as much as you possibly can.

Graham Littlehale: Maybe just as a closing thought, if Endurance is successful in decades from now, what does the world look like? How does it look different?

Andrew Redd: Yeah, I like to think that we are providing global energy abundance. And what that means to me is we are providing opportunity for the world, opportunity to reshore industry, reshore major industrial processes, enable electrification, enable green technology to take off, and really displacing fossil fuel generation worldwide. And so what I would like to see from Endurance's success is the U.S. West Coast, and Japan, and all of the Pacific Islands should be 100% renewable energy.

And then beyond that, as electricity cost gets cheaper, the industrial processes that rely on fossil fuel burning that kind of make up the rest of the energy mix of the world, those need to start transitioning over to electricity and to green sources. And so the way I see it is the cheaper we can make this power, the better we can be for the world, the more change we can create.

Graham Littlehale: Well, Andrew, thanks for coming by and joining us for this.

Andrew Redd: Thanks for having me!


Authors

  • Graham Littlehale

    Partner

Tags

  • Global Resilience

Share