How Sam Cossman Makes a Living Jumping into Volcanoes

The idealistic entrepreneur turns wild experiences into viral videos into actual science into a going business concern.
Sam Cossman

The first time Sam Cossman went into the volcano, it was just for fun.

The 33-year-old San Franciscan was still hungover from the failure of a startup, Qwake, which he had launched to create the kinds of unique, life-affirming adventures that improve the self and (ideally) the planet—the ones that he himself craved and that young, successful people around him in Silicon Valley were increasingly seeking out to photograph, post on Instagram, and discuss at Soul Cycle. Who wants to surf a Nicaraguan break or climb any old fourteener when you can scramble up a redwood alongside a working arborist or follow endangered birds of prey into the best thermals in a paraglider, the latter being a new pursuit called para-hawking that Sam had really highs hopes for.

As it turned out, there wasn’t a profitable business in this, and Cossman was left only with great ideas, one of which he couldn’t get out of his mind. He’d seen photos of this insane volcano known as the Marum Crater, on the distant Pacific isle of Vanuatu, and he was dying to go see it. But not just see it—Cossman wanted to go into it, to get as close to its burbling lava lake, which cast off sulphuric acid and 700 degree heat, as possible. So when an erstwhile client told him that there was an opening to join a BBC documentary crew headed for that very crater, Cossman jumped at the opportunity, canceling a long-planned trip to see gorillas in the Congo with the “extremely loving, supportive girlfriend” who understood this radical, last-minute swerve and is still loving and supporting him today.

Once atop the crater, Cossman put on a shiny, silver industrial proximity suit—designed to withstand temperatures up to 3,000 degrees—and rappelled in, having no real idea what he was getting himself into. He spent a few hours down there, “running around like crazy with my GoPro,” capturing piles of footage of this roiling inferno, a hole in the earth filled with molten rock so angry and orange that it looks like CGI on his clips.

On the flight back to San Francisco, Cossman cut that footage into a short film, added a soaring mix of strings and choir titled “Epic Score,” and posted the video, “Volcano Diver,” to YouTube the minute he had Wi-Fi, thinking only that he was sharing this profound experience with friends and family and Facebook.

The next morning, Good Morning America called.

The video had blown up. (At last check it had 4.7 million views.)

Cossman, who has a finance degree from the University of Georgia, returned to his day job in business development for a med-tech company “as if nothing happened,” but the phone kept ringing—Red Bull, Discovery Channel, and National Geographic all wondered what else this intrepid explorer wanted to do. He was “perfectly satisfied” in this good job that paid well—providing him the wherewithal to buy a BMW, a Ducati, and a quarter-share in a sailboat. “But it wasn’t my life’s work,” he says, three and a half years later. “I had been trying to inspire young people to get out there and see the world by taking one individual one at a time to a unique place, which was obviously not scalable. Suddenly, I made a video that allowed me to do that with a million people at a time.”

This, he thought, was a thing he could build on.

“Maybe it was that people were intrigued by a human being going to a place where a human shouldn't go. Maybe people were intrigued by the fact that someone could have died. Maybe people were intrigued by the fact that there was this incredible force of nature that is so new and unknown. Maybe it as just the majesty of the planet,” he says, still uncertain of the magic he conjured. “Those were all things that resonated with me. I didn't know what it was that had resonated with people. But I figured there's only one way to find out and that's to do it again.”

Sam Cossman has tapped into a form of promotion—exploration that drives the innovation, as he would say—that is equal parts earnest, glitzy, and sly.Carlos Chavarria for WIRED

One impromptu descent into the fiery maw of Mother Earth, naturally, begets another. Cossman followed his first Vanuatu trip with a second one, this time with some sponsors and better equipment, including VR cameras, biometric sensors, and a small flock of drones donated by DJI, most of which got cooked to death, but not before capturing enough footage to create the first-ever 3-D map of a volcano’s crater.

Cossman brought a scientist to emphasize that this was for real and not just another volcano LARPing adventure: Harvard microbiologist Jeffrey Marlow, a NASA consultant and extremophiles specialist who borrowed a prototype Mars 2020 Rover tool he’d built to scan the microbial communities on freshly hardened lava—essentially the newest earth on Earth—and watch how life begins in real time.

The video from this expedition (titled “The Fire Within”) is much slicker than the clip from Cossman’s first trip. It features beautiful high-definition shots captured by the drones, intercut with shots of jubilant children dancing and aerial footage of electric green jungles, plus another epic score and Cossman’s voice, explaining why he had taken a plane, a river canoe, and a helicopter to come back to this remote volcano—“a place that’s forever changed my life in more ways that I can imagine”—and why you are feeling verklempt and a little like a failure in life while watching it on YouTube over lunch. “I’ve always been driven by curiosity and intrigued by the unknown,” Cossman says, as a drone tracks him paddling through a canyon, shirtless. “From a young age, my father always told me, ‘Find a job you love and you’ll never work a day in your life.’”

The video didn’t have the viral impact of the first one. Still, it had the desired effect, raising Cossman’s profile in the media and garnering attention from numerous magazines and websites, including WIRED. He had tapped into a form of promotion that was equal parts earnest, glitzy, and sly. And he could tell he was onto something: “I started thinking about this concept of exploration being the engine that drives the innovation.”

He says this on a sunny day in December of last year, sitting on a bench in San Francisco’s Alta Plaza Park, one arm resting on the helmet he wears when riding his limited-edition matte green Ducati.

Cossman revived the name Qwake from his first failed startup—it was a reference to a life-changing experience he had building a school in Haiti after the earthquake—for this second, as-yet-undefined venture. Four key words kept running through his head: science, tech, exploration, impact. “That got my wheels turning around the concept of facilitating science to be done remotely in places that would otherwise be seen as very dangerous,” he says. “We live in this amazing world that’s rapidly changing. There's so much about those changes that we don't understand, and getting to the core of understanding them is the only way that we'll be able to solve some of the problems that are going to arise as a result.”

The idea wasn’t just to visit volcanoes but to unlock their secrets, using technology. Ditto lightning or sandstorms. If a place seemed dangerous, photogenic, and worthy of some collision of science and technology and social media, Cossman wanted to go there, ideally with American brands helping him pay for it in exchange for the exposure. “I saw it as an opportunity to leverage these cutting edge technologies here in Silicon Valley—the technologies we're using for self-driving cars and drones and genetics and all these other massive profit generating industries—for science,” he says. “To tap into that, bring it over and then create content around it.”

Somewhere in this evolving Qwake ethos was a business plan, Cossman thought, but he had no idea yet how to turn it into a profitable venture—which isn’t the kind of thing that bothers him. Cossman is a specialist in taking leaps, literally and figuratively; he consistently puts his cart in front of a horse that isn’t there and somehow the horse always shows up, just in time.

Also, he has good luck.

After being blinded by smoke and gas during his volcano dives, Cossman had an idea for a firefighter's mask mounted with an augmented reality computer vision system. (This is the fourth iteration of a C-THRU prototype.)Sam Cossman/Qwake Technologies

For instance: Shortly after the second volcano trip, Cossman was walking to a meeting in Manhattan when a random email arrived from a man in Nicaragua who’d seen his videos and wondered if the flaming hellscape that had just erupted in his literal backyard, behind the family’s bed and breakfast, was a lava lake that might be of interest.

Cossman, who by this point received many tips from volcano enthusiasts, asked for proof and got, in return, a shocking video. “I had never seen anything like it,” he says. “It was essentially a raging lateral waterfall of lava,” jetting out the side of a mountain just outside Masaya, a city of 358,000 people.

This got his attention. Lava lakes inevitably do, considering that there are only seven on the planet. Cossman began scouring the internet for scientists and bureaucrats in Nicaragua who might be able to tell him more, or perhaps even help him get there. He heard nothing back for weeks, until one day his Twitter lit up with tags.

Somehow, someway, Cossman’s messages had reached the office of Nicaragua’s president, Daniel Ortega, who had apparently announced to his nation that Famous Internet Volcano Guy Sam Cossman would be coming to study this unsettling new fire mountain outside town. “It became a thing instantly,” Cossman says. “Before Nicaragua and I ever had a chance to communicate about this thing.”

Ortega invited Cossman to visit, and when he arrived in Managua he was greeted by a sea of journalists and then taken to a military base, where various scientists and generals shook his hand and pointed eagerly at maps. “They assumed that I had an elaborate plan that I was there to implement, when in actuality all I had proposed was wanting to come and better understand what was happening and explore the possibility of maybe putting a project together,” he says. “There was no project yet.”

Cossman was flown over the volcano in Ortega’s personal helicopter, alongside scientists who helped explain to him what had occurred. “It was an extremely cool experience, but I have to say, by the end of it, I was starting to feel a bit like an impostor,” he says. “I had come to this place without a plan—yet somehow this was all happening, and I kind of catalyzed this movement.”

He didn’t mention that he was really just a guy with big dreams who did biz dev for a med-tech device company.

Cossman doesn’t lack confidence. He is not easily flapped. And when he got home, he leapt into action. He talked extensively with a PhD who had studied this particular volcano for years and was struck by an increasingly obvious thing: There was a lot more to learn about volcanoes, and the tools he’d taken to Vanuatu—as well as others he could get—might help unlock the secrets of this “massive, complex, living, breathing, system.”

What if, he wondered, you could create an “early warning system” for eruptions, to help improve the lives of the 800 million people who live in the shadow of Earth’s 1,571 active volcanos?

Big data, machine learning, AI, and sensor networks, he thought, could be combined into a holistic system, potentially revealing clues. “And perhaps, if we had some foundational tool that could aggregate all this information and look for patterns that could be something really useful,” he says. Maybe you could connect a volcano and its vitals to the internet.

That idea came out of a meeting Cossman had with Hugo Nordell, a Swedish engineer who then lived in the Bay Area and who had been casually consulting for this new, emerging iteration of Qwake as a technology/exploration company. Meetings between Cossman and “subject matter experts” who seem useful to him tend to follow a pattern; they start innocently, as conversations over coffee, and conclude with the other person as a consultant, offering time and expertise for some future promise of, well, who knows? “He’s incredibly good at pulling people into his reality,” Nordell says.

That’s how he too ended up at the bottom of a volcano.

“Sam, being the pure-blooded optimist that he is, took my word for it that [wiring a live volcano] is possible,” Nordell says. “He created a pitch deck and put my name on it as chief technologist, saying that our team has already proven that this works, and we just need funding.”

Two days later, Cossman called and asked Nordell to fly to New York with him to pitch General Electric.

“I said, ‘What for?’” Nordell says. “He said, ‘Well, I may have pitched GE that you know how to connect the volcano to the internet.’”

Friendships have ended over less presumptuous actions. Nordell rolled with it, helped Cossman win the pitch, and is now a Quake partner.

Six months later Cossman returned to Nicaragua with Nordell to lead a 60-person expedition that included volcanologists, data scientists, climbers, riggers, and, as mission director and team physician, the retired astronaut Scott Parazynski—all of it funded by GE as a marketing vehicle for Predix, the company’s internet-of-things software platform.

Sam Olstein, GE’s global director of innovation, did not know, when he sold the project to his bosses, that Cossman wasn’t 100 percent sure he could pull it off. (Among the things that worried Nordell, who has a day job working for Swedish industrial giant Sandvik1, was how they were going to power a network of sensors in the mouth of an active volcano. “This is not a small problem,” he says.) Olstein had met Cossman a few times at conferences and considered him to be a “next generation Indiana Jones” who actually understood technology. GE was struggling with how to effectively pitch big data applications to corporations, especially as it transitioned into being an increasingly digital company. "You can get bogged down with corporate jargon and a lot of complicated processes and numbers,” Olstein says.

And here was Cossman, offering to deploy GE’s fledgling digital platform in one of the harshest and most complicated environments on earth. If Predix could work for a volcano, it can sure as hell work for your steel mill.

For two months, Cossman and the team traveled in and out of the caldera, installing sensors and using the world’s first zip-line into an active volcano. Cossman promised GE he would do his best to garner viral attention during the trip—that, more than any science, was what the company wanted to generate. He hosted the third-ever Snapchat live story (the first two being a team reaching the summit of Mount Everest and a day in the life of the International Space Station), and did a Facebook live broadcast as he and Parazynski zipped from the top of the volcano down to the lava lake one afternoon. Nearly half a million people tuned in.

Cossman's volcano dives result in some real science; he's also entertaining the idea of doing another volcano stunt as viral marketing for a Nicaraguan rum.Conor Toumarkine/ Qwake

Wearing a shiny tinfoil suit, Cossman stepped toward the edge of the rim around the lake. The ambient air was 500 degrees and periodically, without warning, vents in the wall would belch 800-degree gas, destroying visibility. “It was an extremely inhospitable, dangerous place to be,” Cossman says. Even Parazynski, a man who has been to the summit of Everest and has completed more than 47 hours of spacewalks, was nervous: “As I told Sam, this was the most dangerous, craziest thing I’ve ever done.”

If you listen back to the Facebook audio, there’s a clicking sound, which indicates that Cossman’s oxygen tank was nearly exhausted. (He had a respirator as backup.) “Between the sound of my air running out and having a section below me basically shear off like a calving glacier into this crazy fiery abyss, it was a very intense moment,” he says.

And just then, as “people were staring at this massive molten lake of fire”—the feed went dead. Kaput.

“It was a technical glitch,” Cossman says, with an impish grin. “It was not designed to be a cliffhanger.”

He shrugs.

“Anyway, that was the first real example of how we can blend exploration and technology for good, in partnership with a brand, as a marketing effort, to fund science.”

The most important thing to come out of the Masaya volcano trip wasn’t the jazzy interactive website that GE built around the expedition—known as Digital Volcano—or even the 50,000 data points, collected by the network of ruggedized sensors installed on the mountain, that captured data for more than a month before succumbing to toxic gases and temperatures ranging from the boiling point of lead to the average daily temperature on the surface of Venus. The most important thing to come out of the trip was an augmented reality fire helmet.

Cossman noticed during all three of his volcano trips that there were moments when smoke and steam and gas essentially blinded him—a disconcerting experience when you’re standing on the precipice of hell. Wouldn’t it be cool, he thought, if he could create a tool to fix that problem—to provide sight when blinded by smoke? That would be something with real-world significance, with a target market that could use the help: fire fighters.

Cossman looked around to see if anyone else was working on such an idea, and found C-THRU, a “smoke-diving helmet” created by a Turkish industrial designer and UX/UI specialist named Omer Haciomeroglu. The helmet, he learned, was just a concept. It didn’t exist, yet. But the idea was almost exactly what Cossman had in mind—to rig a fire helmet with cameras and other sensors to provide augmented reality that gives a masked firefighter “sight," even when there’s no actual visibility.

Cossman sent Haciomeroglu an email. Two years later, they are two of the four partners of Qwake Technologies, the applied tech commercialization division spun out to make and sell C-THRU and other products dreamed up while zipping into volcanoes or sailing catamarans into lightning storms. (More on that later.)

The helmet needed a brain; for that Cossman recruited John Long, a PhD neuroscientist and computer-vision specialist who works days as a postdoctoral fellow at NYU Langone Medical Center in New York and nights writing algorithms to give firefighters sight in smoke.

Specifically, firefighters wanted “edge-detection,” to be able to see walls and doors and furniture—but any new system would need to work on low power, be robust, and integrate into existing masks. Cossman learned all of this from the consultants he looped in, especially Tom Calvert, a Battalion Chief in Menlo Park, California.

Calvert is one of the more progressive fire department managers in the US, known for integrating drones into the Menlo Park Fire District’s arsenal. “There's not a lot of huge steps forward that we've seen over the past 100-plus years. The drone stuff is a huge step,” Calvert says, from behind the desk of his office. “But the C-THRU technology with edge detection and thermal imaging built into the mask—going from zero visibility to, ‘OK well there's the walls and click there's the heat’—that’s a big leap.”

In three days, Cossman is leaving for a month in Africa to do research for yet another of his big projects—an anti-poaching effort called Amplify—and he’s in Silicon Valley crossing items off of his action list before boarding a plane for the Congo, where he’ll finally take his extremely patient girlfriend to see those gorillas. Cossman spent the morning at Google X, Alphabet’s R&D arm, talking about possible collaborations and then stopped by Calvert’s office to discuss some tests Qwake would do at Menlo’s training facility with a C-THRU prototype in January.

Calvert dreams of a day where C-THRU enables a system where “we can sling data around on scene.” Drones could send footage to the masks or the command center, where the incident commander can see the feeds from various helmet cameras, and use all the assembled data to help his men see ahead, to make better decisions. Most casualties happen during the first moments after entry, Calvert says, when the firefighters are surprised by conditions or when they get lost.

The first version of C-THRU to come to market—whenever the glacial forces of the first-responder market, driven by municipal budgets, allows—will be fairly basic, with just edge detection. But the idea was always a “platform play,” Cossman says. “We have a road map to build a neural network that is taking all that data and building out a real time map that is constantly evolving and changing that allows a commander to provide the actionable intelligence to the guys in the field.” All of this, he concedes, is “a little bit out in the future.”

People often assume that Cossman is rich. He isn’t. He did not invest early in Facebook or buy 7 bitcoins. “I’m not a trust-fund baby, unfortunately,” he says, over a lunch of Mexican salad and French fries at a favorite restaurant near his home on San Francisco’s Russian Hill. “That would be nice.” His father owned a bookstore called Humpus Bumpus. His mom is a neonatal nurse.

“There's maybe a perception that there’s a Peter Pan element to what I'm doing,” he says, looking a little frustrated. “It’s the question of, ‘Who pays for that? How are you able to just go out there and chase your dreams?’”

But: It’s a natural question.

“Yeah, it's a natural question,” he replies. “And I think, at the core of it, it's a certain level of comfort with risk. Being willing to take that leap. Basically, to jump off the cliff and build a parachute on the way down—no pun intended, for our space project,” by which he means Quantum Leap, the most ambitious Qwake venture of all.

Quantum Leap is an emergency crew escape concept: Future space tourists would be able to eject from suborbital flights and survive reentry at “ballistic velocity,” wrapped in an inflatable heat shield. The project is in early stage development now, and Cossman is working with a team led by Christopher Schulz, the chief scientist for hypersonic programs and technologies at Lockheed Martin. “It sounded crazy, challenging, and really fun,” Schulz says, by email, and Lockheed’s Advanced Technology Center agreed to fund a feasibility study, to be sure that, among other things, Cossman—who plans to test the system himself, by jumping from 300,000 feet, ideally within three years—could emerge “unscathed from the fire of reentry.”

Schultz works mostly on giant space systems, and the idea of doing something fast, cheap, and scrappy reminded him of his time as a project manager at Darpa. “Besides,” he says, “who would say no to the idea of designing a manned suborbital space jump system?” His employer’s interest, meanwhile, is in other applications for a thermal blanket concept. Rocket engines are the most expensive part of a launch vehicle, Schultz says, “and recovery and reuse of the engines alone" could cut space lift costs by 30 to 60 percent.

In these unsteady, early days of Qwake, Cossman juggles grants, favors, and small investments, such as the one Lockheed made in the feasibility study. Other factors keeping the lights on include a rent-controlled apartment with roommates and an ability to get people to help him at low or no cost—often on the promise that one of his outlandish ideas might turn into a business. He can already point to C-THRU. As well as Digital Volcano. And even now Quantum Leap. (There are also more overtly commercial plays; during my visit he was entertaining the idea of doing another volcano stunt as viral marketing for a Nicaraguan rum.)

If Sam Cossman ever feels pessimistic, he disguises it well. “I get the door slammed in my face and part of being an entrepreneur is having the resilience to withstand that,” he says.

Over the three days we spent together, the only time he chafed, and then only slightly, was when I implied that one of his underlying motivations seemed to be a quest for adventure. We were talking about Quantum Leap, a venture that—should it actually happen—will require him to bail out of a perfectly good rocket and crash through the atmosphere like a meteorite. “It’s less about doing something no one’s done. It’s not about breaking a record. That’s as far from my motivations as you can get,” he says. “I have zero desire to be adrenaline junkie. I’m not a stuntman. That’s not why I do what I do. This is a purpose-driven idea.”

That idea: to “revolutionize space safety” and to “reduce the cost of operating in space.” What’s more, Cossman says, “I believe that suborbital point-to-point travel will revolutionize transportation” and if lots of people are going to fly up to 200,000 feet on a parabolic arc as Earth rotates beneath them, the operators of these space taxis are going to need methods in place to “provide rapid egress in emergent situations.”

Like any entrepreneur who spends large chunks of time pitching his business to potential funders, Cossman has a quiver of catchy slogans to explain himself and Qwake. “Everything we do is based around a collective desire for impact,” for instance. Or: “We want to advance science and leverage technology solutions to solve real world problems.”

I remind him of what we’re discussing here. According to astronaut Jonathan Clark—arguably the world’s foremost expert on space safety, and a consultant on the Quantum Leap project—there are “a thousand ways to die” when attempting something like a space jump. The one that stuck in my mind was how an uncontrolled spin causes blood to centrifuge up into the head and then boil off. Or something. This might give me pause.

Not Sam. I have no doubt that he would prefer not to die in spectacular fashion while starring in a Qwake venture, but I think it’s not productive in his mind to focus on such bummers. “It intrigues me,” he says. "It’s the call of curiosity to know what it could be like. It’s curiosity that drives my desire.”

Cossman is earnest even when he shouldn’t be, and his optimism is infectious, which explains how he’s able to get so many talented people to join him. He has no problem cold-calling the world’s experts, to ask, “How do you do this thing that no one has ever done before?”Carlos Chavarria for WIRED

Cossman thinks he could be doing the first Quantum Leap tests, using dummies laden with sensors, within 18 months, for less than $1 million. And that in three years, for “less $10 million,” he could get to the point where “I could egress from a rocket powered prototype and survive.”

“Sam is the new generation. He’s the kid that’s dreaming big and I love that—people pursuing dreams at all cost,” Clark says. But he does wonder if Cossman might be underestimating some of the complexities. Spacesuits, for instance, can cost upward of $1 million.

Cossman respects Clark greatly. But he’s more bullish about rapid progress because things are already happening. Recently, he pitched Quantum Leap to NASA’s re-entry team at Langley, where two different projects have already tested and proven the viability of inflatable thermal shields that could protect him while re-entering the Earth’s atmosphere at hypersonic speed. Getting up there is a huge cost, he says. He needs to align with launch partners, like Blue Origins or Space X or Virgin Galactic, all of whom he’s already approached.

Qwake does not have a headquarters in the traditional sense. The company is based, I suppose, in Cossman’s pocket or his messenger bag or wherever his iPhone X happens to be, though he prefers to say that it’s based out there, in the ether.

His three partners in Qwake Tech are in New York, Seattle, and Turkey; the two who work on Amplify are in Los Angeles and Sweden (with consultants in Virginia and Holland); his brand strategist is in New York; and the main Quantum Leap guys are in Los Angeles and Boston. “I’m definitely not the thought leader in every one of these projects,” he admits. “There needs to be that connective tissue to bring it back to the business. At this point in time in the company's development I am the front man for this stuff. But we need money, not a building.”

The Quantum Leap project, for example, dates back to some early conversations with National Geographic, which asked Cossman to pitch audacious ideas—as big or bigger than the volcano descents. Scott Parazynski introduced him to Clark, and both agreed to consult, since Cossman is neither a rocket scientist nor an astronaut. He’s also not an engineer, but he found one of those, too, in Chris Schultz. “I’m not an authority on much,” Cossman says. “But I do feel like I’m an authority in understanding the power of bringing lots of heads together around a unified goal.”

When he is in San Francisco, Cossman spends much of his time in the three-bedroom penthouse at the top of an old Victorian in Russian Hill where he has lived for 10 years, with a rotating cast of roommates. Currently, that means his girlfriend, who does corporate social outreach for a tech company, and a European app developer who, like Cossman, works from a laptop in the shared family room, which has spectacular 270 degree views of the city and the Bay. It is living room as co-working space (“and sometime rapid prototyping studio”) decorated in the Persian style of an ex, which gives the place a bit of an opium den vibe. If you’d like to work on a pile of silk pillows under a little tent, you can do that, too.

Cossman’s bedroom, in the back, is dominated by a monolithic four-poster canopy bed from Indonesia. “I bought that a long time ago,” he says, when he notices me knocking on one of its intricately carved posts, as thick as a middle-aged tree. Biographies of Elon Musk and Richard Branson sit on his nightstand, as does a piece of the zip-line from the Nicaraguan volcano, rusty from the corrosive air; a python skin the length of the wall hangs above his bathroom door. He got it during the year he spent traveling the world alone, after working 12 months of doubles at a P.F. Chang's to cover the trip. “We ate that snake,” he says. “That was a different life.”

In addition to the best apartment in the five story building, Cossman somehow got the one-car garage, which houses his SUV, motorcycle, a sea kayak, and racks of supplies from past expeditions. There are 12 unopened “sacrificial drones” from Nicaragua, numerous zip-line components, and two boxes that contain pieces of the “lava gun” he and Omer designed to capture samples of molten lava, something no one had ever tried before. They started with an off-the-shelf tactical line thrower and modified a special graphite tip by boring holes that, they hoped, lava would dribble into and harden. (Graphite will vaporize at 6,000 degrees, but the lava lake is only 2,500.)

“We didn’t run into issues with the fire,” Cossman says. “It was the retrieval process.” They neglected to bring a winch and tried to retract the line by hand over jagged volcanic rocks. It jammed. “We learned many lessons.”

Cossman is an extremely pleasant person to spend time with; he is earnest even when he shouldn’t be, and his optimism is infectious, which explains how he’s able to get so many talented people to join him in projects that often have no obvious funding sources or commercial futures. Cossman has no problem cold-calling the world’s experts, to ask, “How do you do this thing that no one has ever done before?” This has made him many unusual friends, enabling him to “build up this amazing Rolodex of scientists, tinkerers, and hackers” who are willing to heed his call, if and when he finds the right project, plus the time and money.

It can sometimes seem as if Cossman’s day is broken into one thousand tiny meetings, conducted via every possible method of modern communication. He pings between Slack, email, text, and phone, juggling discussions among the core group of Qwake “misfits,” and the outer rings of his solar system, populated by a growing array of what he calls “SMEs” or subject matter experts, most of whom he knows because he cold-called them one day and asked for help. “If all this fails, Sam could teach a whole class on how to write an email,” C-THRU engineer John Long says. “That guy’s ability to write an email asking somebody for something while making them feel good about it is amazing. He’s very good at expanding the pie.”

In 2015, Cossman visited Kava Khan ("the Michael Jordan of eagle hunting") in Mongolia. He wanted to put a GPS tracker, camera, and sensors on an eagle to study things like “the flight metrics of the bird." He had planned to do it on his own, but at the last minute Red Bull came with a video team to film the expedition.Justin Bastien

Last year, Cossman spent two weeks on a boat in Bora Bora with his girlfriend, where he met the professional snowboarder Travis Rice and his filmmaking partner, Justin “Chip” Taylor. Cossman befriended them both and ultimately convinced Taylor to make a trailer for Planet Chaos, the show he was then developing along with Scott Parazynski for National Geographic.

The show would “stage expeditions leveraging technology to push the boundaries of human knowledge,” with every expedition thrusting Cossman into a complex, unfriendly environment to try and understand some natural phenomena that science had yet to tackle.

He whips a laptop out of his bag to show me the trailer. Like all of Qwake’s media, it is pretty, a mix of hi-def footage and motion graphics designed by a guy he found on Craigslist who was just hired to do video for the Golden State Warriors. “I’m so stoked for him,” Cossman says. (The guy will keep moonlighting for Qwake on the side.)

Among the expeditions Cossman pitched was one calling for him to rappel deep into holes in the Greenland icecap to deploy sensors and study the transition zone between ice and water, to better understand how rapidly the ice sheet was disappearing. For another, Qwake would modify an experimental catamaran to add a faraday cage and a sensor array and sail it out into the middle of Venezuela’s Lake Maracaibo, which gets more lightning strikes than any other place on Earth—more than 40,000 a night in peak season. He wanted to dive into an Indonesian acid lake that gives off sulfur gas “the pH of battery acid”; descend 1,000 feet into the “world’s deepest blue hole,” recently discovered in the South China Sea; and drive into the teeth of a Saharan haboob to study the movement of pathogens over long distances.

“This is a project I’m developing right now,” he says, pausing on a digital illustration of a modified Humvee leading a swarm of drones just ahead of a vast and menacing cloud of sand that could easily be the poster for a disaster porn film starring the Rock. It’s a project Cossman calls “Decoding Sandstorms.” He says he’s already secured the tentative support of “the President of Chad,” whose country is plagued by the storms, through a partnership with a renowned German geologist. “I owe them an email,” he says, and clicks over to a message forwarded from the geologist that assesses the relative dangers of going into what is now a war zone.

All of these projects were pitched for Planet Chaos, but are now part of what he calls “a Chinese menu of sorts” that he shops to partners. This one, Cossman says, would be “perfect for an automaker.”

When I arrived in San Francisco, on a Monday, Cossman was working on the Amplify concept, which had evolved from a vague notion that sensors and Big Data could be really useful to detect and predict poaching. He has a beautiful pitch deck, a compelling one-minute trailer with soaring music and slick infographics, and would soon be testing prototypes and chasing sponsors.

One of Cossman’s partners in the project is Brad Halsey, a PhD chemist, tinkerer, and entrepreneur who quit a perfectly good think tank job to create and run a rapid prototyping unit for the US Army in Iraq. (“I was an embedded geek,” he says.) Now he has his own engineering start-up, Building Momentum. Cossman first called for help with Planet Chaos, when he “was looking to go kill himself in 50 different natural environments,” Halsey recalls. He wanted a technical specialist to help him execute complicated missions using deployed technology and, when necessary, to “throw the bullshit flag."

Halsey liked Cossman immediately. He just had a gut feeling. “He’s so energetic that you’re like, ‘He can’t actually be for real.’ He’s almost an avatar of something that’s real,” Halsey says. When he told his team about the conversation, there was a little hesitation. “Some of them were like, ‘Is this guy full of shit?’ I said that he has the foresight to ask us to help, so I’m giving him the benefit of the doubt that he knows his strengths and weaknesses.”

This ability—"to realize the gaps and start demonstratively filling them with people who aren’t him”—is impressive, Halsey says. Cossman is “the connector,” he says, “the one who puts all the pieces in the right spots and adds the motivation for that whole raft of people to paddle forward.” Halsey agreed to participate in the anti-rhino poaching project if Qwake could get the funding to build sensors and deploy a test network on the ground in Africa.

Wednesday morning, Cossman texted me to say that he’d just gotten off the phone with Jan Kees Schakel, the founder of a Dutch startup called Sensing Clues, and that by the end of their two-hour call they had decided to pursue a joint venture. By Friday, he’d looped in Halsey and Building Momentum and sketched out a structure for how Amplify would become a real thing.

The first version of Amplify will be fairly simple—a sensor network that listens for gunshots and alerts rangers when they occur with the approximate location. But like C-THRU, Amplify is to be iterative. Once the network is in place, you build on it. Sensing Clues had the machine learning and a dataset that could, already, identify more than 100 different types of gun shot using inexpensive sensors and a low-power network. Building Momentum would ruggedize those sensors, optimize the system, and figure out how to install and link it in the bush, assuming, Halsey says, that it “will be a total disaster the first time. It always is.”

Somehow, in two calls over three days, Cossman had recruited two entrepreneurs in two countries who’d never met to collaborate with Qwake on Amplify, and not just that—they had sketched a rough plan and were “moving rapidly” to make a prototype and pitch to Google X, ideally by March.

Through a friend, Cossman had also met the engagement manager for A3, the advanced technology arm of Airbus, based in Silicon Valley. That contact, who found his “energy and will to do good contagious,” passed him on to the Europe-based Geo Intelligence team, which gave Qwake free temporary access to OneAtlas, its space imagery database.

He flies out Saturday for a month-long recon trip to Congo, Rwanda, Kenya, and South Africa.

There, to sweeten the package, Cossman will capture compelling footage of himself shadowing rangers and nuzzling rhino babies, having already used his many newfound, sometimes improbable connections to arrange meetings with park rangers, anti-poaching activists, government officials in three countries (Congo, Kenya, and South Africa), and, he hoped, famed paleoanthropologist Richard Leakey. Cossman’s solar system, like the universe, is ever-expanding.

1 Correction appended, 2/28/18, 7:10 PM PST: A previous version of this story misstated the name of this company.


Josh Dean is a correspondent for Outside and a regular contributor to Popular Science, GQ, and Bloomberg Businessweek. His latest book is The Taking of K-129: How the CIA Used Howard Hughes to Steal a Russian Sub in the Most Daring Covert Operation in History.


More WIRED Longreads