Building Boundary-Breaking Balloons

Pushkin. Every day around the world, more than a thousand weather balloons are launched into the sky. The balloons float high up into the atmosphere, sending back information you know, temperature, wind speed, air pressure, et cetera, and then a couple hours later, the balloons pop. This is basically the way weather balloons have worked for decades, and the information the

The first half dozen flights of our balloons were complete and total failures.

When do things start working?

That would be back in June of twenty sixteen. This is the end of my freshman year, and I think our longest flight that year had been fourteen minutes, so shorter than a normal weather balloon.

Buy a lot, Buy a lot.

And also the person who had started this project, who then became my co founder, had just graduated and was about to leave what we thought never to be seen again. And school has ended. We can't even get into the building because our key cards have stopped working, so we have to tailgate someone in to get our balloon launch equipment out. Go out to the middle of nowhere in the central Valley of California, and if this launch didn't succeed, windborne would not exist.

Okay, so I can guess what happened next based on that fact. But tell me what happened.

Yeah. So it got to be about four am, and we're out in the middle of this park, really cold, and the sprinklers turn on. They start slowly swiveling around towards our delicate electronics, and so I have to hurl myself onto the sprinklers, getting soaked, just to save our electronics.

It's like the nerd version of throwing yourself on a grenade exactly.

I really felt like I was in an action movie sequence. Ah good, But that sacrifice really paid off because with that flight, we managed to set the world record for weather balloon endurance.

How long did it stay up?

It stayed up for seventy six hours?

Wow, and your previous best.

Was fourteen minutes exactly.

So you're in college, how do you build weather balloons that are better than whatever the federal government? Noah's launching our you know, state of the art weather balloon.

People. So the first big thing is that some new consumer electronics have come out, in particular, things like micro controllers and lightweight, low power satellite communications, and we realize that we can use these innovations to make a balloon smarter and to control its altitude.

What's a micro controller.

Micro controller is a super small computer that you can put in any device to give it a brain.

And they're cheap.

They're dirt cheap. Anytime you hear the word Internet of things, that means micro controlers.

And nobody else was doing this at the time, Like all the big, you know, well funded weather agencies weren't sticking my micro controllers on their on their weather balloons.

Well, with these big agencies, they are really incentivized to keep things stable because weather is this really important thing that billions of people depend on, and so they're not going to go and say, hey, what crazy experiments can we run.

Their incentive is just like, don't screw it up exactly the way we did it yesterday. Don't screw it up and you won't get fired exactly. And so specifically, what was it that you were doing that no one had done before.

Yeah, So it's a very simple concept that is really hard to make work and practice. The way we control our altitude is when it gets too high up, we vent some of our gas so it has less lift. When we fall down too far, we drop some ballast so it stops falling. But the hard part about this is it has to function at negative seventy degrees celsius. That's the temperature of dry ice. You have to have

So, okay, so you set the record, you stay up for what three days watching it? How far did it go?

It flew all the way across the country and landed out in the Atlantic Ocean. But the real reason why we decided to form a company around this was because we realized the impact this could have. We got into this from the engineering side. The lands are fun, but we realized, wait a secd eighty five percent of the Earth's atmosphere is invisible to humanity. Weather is this crazy,

I mean, I buy that the stakes are high. I buy that it could be very helpful, but like, why wouldn't anybody else do it?

Yeah, a lot of people are trying to improve the weather forecast. But what's unique about Windborne is the combination of our balloon technology with our AI weather forecasts, because really those are the two levers to pull to increase the accuracy of a weather forecast.

You mentioned that eighty five percent of the world lacks good weather data, which was surprising to me the first time I heard it. But I would think, like, get satellites do it. That's kind of I guess my naive ish thought. It's like, I know there's a lot of satellites looking down at the Earth all the time.

I'm actually a big fan of the space industry. I used to work at SpaceX. But the problem is the laws of physics fundamental limit what satellites can measure. Take for example, pressure. It turns out pressure is extremely important for predicting the weather because it determines where the winds are blowing, how the weather systems are moving. But a satellite fundamentally cannot measure pressure from space because it's not in the atmosphere. It can't see what's going on.

So you start the company you're good at making weather balloons that go up and stay up for a long time, and that you can steer. I also read that you launched thinking that you could collect data for a tenth the cost of existing alternatives, which sounds compelling, but as I understand was not compelling enough. Is that right?

It is. We're now at one hundred and fifty times more data per dollar than alternatives.

Okay, was there some reason one tenth the cost was not cheap enough.

Yeah. When we started the company, we thought it would be cheap enough. But it turns out when you really look at the scale of weather data collection, you need to really understand the atmosphere. You want ten thousand balloons aloft concurrently, that's a balloon every sixty miles in the atmosphere. And in order to get to that level of scale, it's the difference between your company spending one hundred million dollars a year and spending a billion dollars a year. One was a lot more in reach.

How do you drive down the cost to one one hundred and fiftieth the cost? I mean, I'm sure it's many many, many incremental you know, efficiency gains. But like, what's an example of one.

Yeah, Well, one of the big pieces is improving the software that flies on it so that the balloon can fly for longer. I talked about in the student group that first flight lasting seventy six hours. Now our longest flight can fly for over four days.

Wow.

And the longer you fly, well, the hardware costs stays the same, so that means you're collecting a lot more data.

Yeah. Yeah. So basically, make the balloon stamp for longer is the fundamental way that you make it cheaper exactly, And how do you make it fly longer?

Yeah, Well, one of the big things is improving the software to better decide when do you vent gas, when do you drop ballast or really one of the other big things is just making everything smaller and lighter, because the smaller it is, the lighter it is, the less chance of a leak, the less ballast you have to use.

Yeah, that's amazing. So okay, so let's talk about where you are today. Let's talk about sort of how it works. How big are the balloons?

So the balloon is two pieces. There's the envelope that's the bag that holds all parts the balloon part of the balloon, and then there's the main unit, which is the electronics and the ballast.

The stuff exactly, the balloon and the stuff the balloon part.

Of the balloon. The balloon part of the balloon is five and a half meters tall. So that's two and a half three times the height of person, depending on how tall you are.

Yeah, so tall. Yeah, it's a big balloon. And how how big is the stuff? How big is are the sensors in the part that's not the.

Yeah, that stuff is. I guess this is an audio thing, so I would say about this big.

The size of a basketball.

Yeah, it's a roughly the size of a basketball. It weighs just over four pounds, so about the weight of a large duck. Okay, and is kind of long and skinny. It kind of looks like a fish. I think of it as a trout attached to a giant bag.

How many of your balloons are in the sky right now? About?

Yeah, there are a few dozen aloft right now.

Okay.

We launch around one hundred a month and are quickly ramping that up.

So where do you launch your balloons from?

We launch them every day from three continents, South Korea, paloelto New York and Cabo Verde.

Cabo Verde is just off off the west coast of Africa. Is that right?

You got it. One of the reasons why we have launch site in Cabo Verde right now. That's right where a lot of hurricanes are forming, and so the fact that we're collecting data around there is going to be really impactful for better predicting the path of these hurricanes. Huh.

So they start there and then they essentially travel across the Atlantic and into the Americas, and if you can understand what's going on there, ideally you can sort of understand the hurricane where it's going to go in the dream scenario, even just as it's becoming a tropical storm.

Exactly. Yeah, Noah did an analysis of the Impactor data had on the twenty twenty two hurricane season, and it made the forecasts for Hurricane Fiona about twenty percent better.

Now it's time for a few ads. After the ads, Kai and I will talk about AI. Of course, we'll also talk about Winborne's business, what they're selling, and we'll talk about the company's quest to build balloons that can stay in the air for months at a time and make multiple trips around the world. What is your business right now, like, what are you selling and who you're selling it.

To right now. Our business has two pieces to it. We are selling the observations of the atmosphere we collect to the government right now, so that those governments can use this data to improve your weather forecast. And just a couple of weeks ago, we announced our AI based weather model, which is the world's most accurate global weather model bar none.

I feel like you need a little TM when you say the world's most accurate global weather model bar none.

We really do, we really do.

So let's talk about that. Is that claim validated? I mean, I know there was news about that just this year, and congratulations, but also, how do I know that's true? Respectfully?

Yeah, I love that question because anytime a weather company makes a claim, you should look at it closely. So, first off, we've published our results on our website and so you can dig into some of the raw numbers there. Second, we're in the process of submitting to weather bench which is these benchmarks run by Google, which were the previous holders of this record. So we're actively talking with them right now about submitting our results getting them validated into

Let's talk about the model you built. So I mean generally when I talk to people who are working on AI, they talk a lot about data, right, And it seems like to some significant degree the models are kind of commoditized, not quite, but like models are pretty similar, it seems in many settings at least one to the other, and the differentiator often ends up being the data. Is that the case in this instance? I mean, do you think

I think that you're spot on. Data is the real mote here, because what really sets us apart is the ability to have all of this data that no one else has and to use that both for training the model but then also for running the model in real time. Data isn't just about training. It's about essentially giving your model the prompt of what's going to happen next, and you need new data every single day for that.

Just one thing to clarify. So, if you're selling the data to governments, are they not sharing it widely? Like they're paying for the data but they're not. But not everybody can get at the data? Is that why it's a mode for you?

Yeah, so what they're using the data we sell them for is putting it into conventional physics based weather model that they then release. So they can't redistribute the data itself, but the general public can benefit from its use in these weather models.

And so just to distinguish and this is a distinction that's not particular to your company, but you mentioned the sort of traditional physics based weather models. I mean, I think it's worth spending just a moment here to distinguish, right, like between the classic weather model, you know, the sort of pre AI weather model, and the AI weather model. Like, what's the basic difference between those two?

Yeah, So a conventional physics based weather model, it takes the initial state of the atmosphere and then runs a bunch of fluid dynamics on it to simulate what's going to happen to all of these different fluids.

There's a sort of kind of classic sort of feels like kind of nineteenth century deterministic. Give me the initial condition and I'll tell you what's going to happen in the future. It's that right, it's sort of rules based.

Exactly exactly, and that is nice because you can see the exact physics that is being used, but you need compute clusters. They cost hundreds of millions of dollars in order to run this because the atmosphere is so big. And by contrast, a AI based weather model can run on just a gaming laptop.

Huh.

And what it does, by contrast, is effectively picking out statistics.

I mean, it's just it's machine learning. I presume it's pattern matching exactly.

It's our model is a transformer based architecture, the same thing that powers chat GPT.

Our A models in general better at this point than the physics based traditional models.

It depends on the use case. But if you're talking about hurricanes, yeah, we ran a case study on Hurricane Ian and our model would have predicted landfall by two hundred kilometers closer.

I mean, that's another one where like I'm intrigued by what you said, but I'm very eager for I'm not eager for this hurricane season to come, but like it will resonate more with me when you do that prospectively, right, of course, I mean, will you this year be predicting where hurricanes are going to make landfall?

We will be predicting that for public safety reasons. We aren't going to be you won't tell me.

Yeah you will, but you won't tell me we.

Will know a hurricane. Okay, but fair, you of course don't want to say, hey, don't listen to Noah about these evacuation orders.

No, no, fair that that is very responsible. So basically, you're selling data. You are likely soon to be selling forecasts. That's sort of the business. On the technical side, what is the frontier? What are you trying to figure out that you haven't figured out yet?

Yeah, well, one of the big areas for innovation is on the AI modeling side. We were kind of surprised that our models did as well as they did because there are a lot of quite frankly obvious things that we haven't done, things like just increasing the amount of compute we're using to train these models. We use something like a fifteenth as much as Google did. So what happens when you train it for longer?

I mean, I'll say, on the AI side, it's not obvious that it's optimal for you to be doing the AI as well. Right, there's a universe where you are optimizing the balloons and the data, and then someone else like Google, who has all the computers and all the AI knows what to do with doing the AI side.

Of it, right, Yes, and no, I think that where that falls apart is coupling our data with the models much more closely. In that there's so much to be done in terms of figuring out how to better take advantage of our data into particular and also things like saying, based on this weather model, where should we be flying our balloons to improve the forecast? And we're already using our own AI weather forecasts to do that flight plan optimization to figure out where our balloons are going to fly.

That's a good feedback loop.

There's also a lot to do on the balloon side of things. I wish I could tell you about our project that will increase flight time by another factor.

Of ten, another factor of ten.

Yeah, so what what are you at now?

What's the like median flight time?

Medium flight time is seven to ten days depending on how we're targeting it.

And so you think you're going to get to ninety days. Yep, we around the world in eighty days. Yeah, you say you wish you could tell me like you feel like you're about to do it. Yeah, tell me, tell me without telling me what you can't tell me.

Okay, I'll do my best. So we have had some very successful flight tests where we have demonstrated the ability to fly without using any finite resources.

So what what is the key finite resource? The the lift gas, the what is it helium or hydrogen?

What you used to keep Our balloons are compatible with both helium and hydrogen, So which we use is dependent on the country.

When you say flying without any finite resource, that's what I think of, yep.

Okay, yeah, that and ballast.

Oh right, So so you want to you can? I should? I guess.

I mean, there's heat there, leave it there and just say.

If you get that, could you fly forever?

That the dream someday, someday we're we're definitely ninety.

Days for a weather balloon is sort of forever, right.

It really is. And so, yeah, we've had some very successful flight tests.

What's the longest flight you've had to this point?

Forty days?

Forty days? Okay?

Long?

Yeah?

Long? And that was without using this technology, was that like a fluke. It was definitely on the side of the bell curve. But we've had multiple month long flights.

What why do your balloons usually what do you say fall? Crash sounds a little extreme crash fall.

Yeah, it's that they run out of ballast and lifting gas, and so these finite resources have been used up. You don't have a more ballast to drop, you don't have a more gas event, and so you can no longer control your altitude.

What happens when it falls crashes comes down.

So by the time it falls, it has used up all of the ballast. So instead of weighing four pounds, it weighs about two hundred grams.

Two hundred grams is what half a pound?

Ish? Yes, I'd looked it up. Zero point four to four pounds.

Okay. By the way, when you're getting rid of ballast, is it just like sand or something? You just like sprinkle sand out exactly? Okay, So it's half a pound, so it wouldn't hurt if it hit me on the head? Is that is that part of the reason half a pound is important? Would it if it hit me on that?

Is it? Anybody on that it's never hit anybody. We do landing simulations and control where it lands to direct it towards unpopulated areas. But remember it also has this envelope attached to it, which acts like a parachute, so it actually falls very slowly. It's very light and not an issue.

And then what happens. Then you have this big deflated balloon sitting on the ground or floating in the ocean, and is it just what happens to you?

Yeah? So I love that question because it's a chance to talk about how our balloons can reduce the amount of waste going out into the world compared to the half a million that are launched every year, of which only a fifth are covered.

Just get answer directly. I appreciate that you want to contextualize it, but like, first, let's talk about what happens with your balloons and then feel free to provide the bigger.

Cart sounds good, sounds good? Yeah, yeah, So our balloons are out in the world, and we recover as many of them as we can. In the long term, we're aiming to recover essentially all of them because we'll direct them to specific landing sites.

I mean, the ocean has got to be tough to write. I presume when they fall in the ocean, for the most part, you don't recover them.

Is that right, exactly? And that's one of the reasons why improving endurance is so exciting, because if you circumnavigate multiple times, well, on the last leg, you bring it down in a field, have somebody drive around and collect all of them. They are about half million weather balloons launched each year, and only a fifth of those are recovered. So when each balloon flies for only two hours, well, you have to launch a lot more balloons to collect

Right, So there will actually be fewer balloons going into the world if you succeed.

Exactly.

So, you're at this point where you've learned a lot. You sort of feels like you're kind of on the precipice of a lot more. And I want to talk about sort of two futures. Right. One is a future where where your company doesn't succeed for any number of reasons with which you're truly more familiar than I. And then the other is if you do succeed. Right, So, in the sort of sad version where you don't succeed, what are some reasons it might not work out?

Yeah, I think that probably the biggest reason is that scaling up hardware is hard. We need to increase manufacturing by a factor of one hundred effectively, and.

So it's scaling up that final assembly by a factor of one hundred is complicated and hard and just kind of classic hard building. A business gets more capital intensive exactly, and the money before you can get the money exactly. Okay, So that's an easy to imagine sad outcome. Let's talk about the happy outcome. Like it works. You scale up your balloons, stay up for months at a time. What's the world look like? What are you doing in that scenario.

We want everybody to see twice as far into the future when it comes to weather. So making the ten day forecast is accurate is the current five day forecast. Making the twenty day forecast as accurate as the ten day forecast, so we want people to see twice as

We'll be back in a minute with the lightning round. Okay, let's do the lightning round. Would you rather have it be too hot or too cold?

Too cold any day?

Okay? The next one is multiple choice. What is your favorite movie prominently featuring balloons? Wizard of Oz around the World in eighty days, up the Red Balloon, or none of the above.

It's got to be up brilliant.

Movie, Blimps, overrated or underrated?

Ooh, so here, I've got to make a distinction between blimps and Zeppelins because I actually used to work at a zeppelin company.

Is that the Larry Page Zeppelin company? It sure is perhaps the only zeppelin company. Yeah, and so is the distinction that a zeppelin has a rigid frame?

Got it in? Won?

Okay? So how about this. Let me reformulate it to see if I get it this time. Airships overrated or underrated?

Underrated? I thought that zeppelins are so cool. I don't have a great mission driven answer here other than you know, I think it's amazing. I read too many books growing up which prominently featured airships.

Isn't the isn't the Larry Page airship company dreams some kind of cargo, like the idea that, for like really heavy things, giant airships would be an efficient way to move cargo.

Full disclosure, It's been a long time since I worked there, so I can't speak to that company in particular. But yeah, faster deliveries than a ship, more efficient deliveries than a plane.

Aha, It's like a niche an Until what's something besides the weather that AI will be really good at predicting?

Oh? Great question, let's see. Uh. I don't have a good answer off the top of my head. I think that it will be really interesting for economics in terms of taking short term data, various real time indicators, and making predictions about what the final readings are going to be.

I'm sure a lot of very smart people are working on that and they may in fact already have good models and they're not telling us.

I think that's likely.

Is there anything else you want to say?

I think that covers it. Yeah, great to be on the show.

Thanks for your time. It was lovely to talk with you. Good luck.

Thanks and if you ever are in Paloelto, you're welcome to come see a balloon launch.

Great. So do you still you launch from where you launch? From that Air Force base or like or just from.

From our part part?

Are they easy? Yeah? From the parking lot, that's cool.

Yeah.

How much space do you need to launch a weather balloon?

Not that much? About fifty by fifty feet. It's really just a matter of making sure there's nothing that the balloon will blow into right after releasing it.

Uh huh. Seems like it'll be fun to see. Yeah, I like a balloon. Kai Marshland is the co founder and chief product officer of Windborne Systems. Today's show was produced by Gabriel Hunter Chang and edited by Lydia Jean Kott, who was engineered by Sarah Buguer. You can email us at problem at Pushkin dot FM. I'm Jacob Goldstein and we'll be back next week with another episode of What's Your Problem,