From SpaceX to Robotics For Good: Keenan Johnson, former Head of Engineering @ Impossible Aerospace

Updated: Jan 20, 2020

"I hope that 2020 will be the year we start to connect some dots. That the automotive industry, medical device industry, the aerospace, and other industries can find a common vocabulary and start establishing common lessons amongst them." - Keenan Johnson

This episode features Keenan Johnson, former Head of Engineering and Founding Engineer at Impossible Aerospace, an American company aiming to end the age of kerosene propulsion and fully electrify all air transport. During the interview, Keenan shared his career experience from contributing to the world's first orbital-class rocket at SpaceX to leading the development of the world's first electric aircraft. He also discussed an exciting new venture on building robotics for impaired kids and environmental education.

Keenan Johnson, former Head of Engineering @ Impossible Aerospace

Keenan got his start in aerospace working with NASA to develop novel techniques for satellite propulsion and performing CPR in microgravity environments. He then moved to SpaceX as a software engineer and mission control operator, where he worked on the world's first orbital-class rocket capable of reuse. He is currently exploring several project ideas involving AI and robotics to help the visually impaired and educate children about renewable energy.

Highlight 1: Turning Falcon 9 to A Reusable Rocket

Highlight 2: The most rewarding moment at SpaceX

Full interview:

Listen to podcast on: Apple Podcast | Spotify | Google Play Music is a content platform dedicated to helping engineers and researchers develop leadership, entrepreneurship, and AI insights to scale their impacts in the new tech era.

Subscribe to our newsletter to stay updated for more inspiring talks:

Full Transcript

1. Why SpaceX?

Margaret Laffan: Good afternoon Keenan and thank you for joining us here.

Keenan Johnson:

Thank you for having me.

Margaret Laffan: You joined SpaceX right after you graduated from the University of Missouri in 2014. Why did you choose SpaceX?

Keenan Johnson:

I went to SpaceX because I wanted to take part in the hardest engineering possible in the world, which was aerospace. I think it's also undeniable that the space exploration industry has done a lot of good for humanity, which was something I was eager to be a part of.

Margaret Laffan: What were you thinking when you were looking at SpaceX specifically? Were you considering the types of opportunities available?

Keenan Johnson:

I didn't really know too much at the time. Nowadays, there are quite a few space startups and companies that exist. At the time, it was basically NASA, the big defense contractors, and SpaceX, which were the new people on the block and nobody was really sure if they were going to last or if the commercial space industry would be real. All I knew was that I wanted to be doing something really interesting. And SpaceX seemed like the most interesting place at the time.

2. SpaceX’s Major Breakthrough in 2013/2014

Margaret Laffan: What was happening during 2013 / 2014 that triggered such growth? And you were there for that period of time?

Keenan Johnson:

Definitely. When I came, they had just flown the fourth launch of the Falcon 9 rocket, which is the big rocket. And they were flying about once every year and a half or so. When I showed up, there was pretty much only one rocket being built at any given time; everyone was focused on “the rocket”. That's what aerospace had been like for NASA, too. It was always like “the rocket”.

That's what everyone in the company worked on. There were no spares, and there wasn't a production line or anything like that. Over the course of the few years that I was there, we went from that to having probably seven or eight rockets in the factory at a time. We could build a rocket in about two weeks when I left, which was an insane period of growth.

I think there are a bunch of factors that contributed to such growth. First of all, the business climate at the time played a large part. Suddenly there were enough customers who had enough faith in SpaceX to buy that many launches. Additionally, the engineering team and the production team were big enough and good enough that they could handle that.

Margaret Laffan: What's the experience like on a launch?

Keenan Johnson:

Going to a rocket launch is the most incredible thing. I know people like to talk a lot about product launches with similar terminology to that of rocket launches, simply because they're so exciting. We were the team that sits in the Mission Control Room and we were wearing the headsets and staring at a bunch of screens, videos, and data.

One of my biggest jobs for the first part of my time at SpaceX was writing all the software that did that. And as part of that, people talk about DevOps engineers. What we were doing was writing the software and then doing the operational side of it. When I started, rocket launches used to be a very long process. The process would start three or four days before the launch because we would take the rockets out and do a test firing as the last check to make sure everything was working. You only get one shot at it. The actual countdowns would start 30 hours before the rockets would go. There was a long period involving generally pretty boring stuff where the tension slowly increases. The last hour or two is the most exhilarating thing ever because more and more stuff starts happening; it starts to get more real. And then comes the moment when the rocket actually lifts off.

Margaret Laffan: At what point did you feel relieved?

Keenan Johnson: When I had gone to see rocket launches when I was little, they would take off and then it was over. When you're actually working there, the relief doesn't happen until you let go of the satellite, that’s the reason you're doing launches: to put something in space. That doesn't happen until between like 10 minutes to maybe an hour after the launch happens. Even if the whole rocket launch goes perfectly, if you don't actually let go of the satellite, then it has no purpose.

I think there were some periods of relief. Once the first stage separates, that was a big relief. There was some relief with each milestone - the stage of separating, and then the fairing coming off to expose the satellite, and then the satellite being launched.

3. Turning Falcon 9 into a Reusable Rocket

Margaret Laffan: When you were working with SpaceX, one of your major contributions was to improve the Falcon 9, so that was looking at converting a single-use rocket over a year into a fully reusable rocket that would launch around 10 times a year. From an engineering perspective, how did you approach this and what was the most challenging part of this process?

Keenan Johnson:

There were some things that you had to decide upon at the very beginning. The first big choice was with regards to the rocket motors, many of which are solid motors which are inherently single-use. Obviously they were not suitable for us. The second thing was that a lot of the liquid fuel rockets that existed at the time could only light the engines once. This makes them more reliable, but not reusable. There were a lot of decisions like that which got me on the front side. That stuff is difficult for sure, but relatively straightforward, but it's so important to have that mindset of: “We’re going to reuse this, that’s going to be hard, and that's okay.”

I think the more challenging aspects ended up being on the software lines. A lot of that was doing testing and running automation to determine things such as whether the valve stilled looked like it was working. Or whether the value still looked fine. We had the most data compared to anybody that had ever flown a rocket before, because we built it using basically off-the-shelf computing technology, so we had all this bandwidth where data, memory and everything else had not been a thing for rockets before. And then we had to think about what were we going to do with all this data.

4. Most Exciting Moment During SpaceX

Margaret Laffan: What was your most exciting moment during your time at SpaceX?

Keenan Johnson: The most exciting moment for me was seeing the first rocket land. Nobody thought that a rocket could re-land vertically in Florida, that was a really key enabler for doing these cheaply. I got to be in Florida to see that landing happen, which was incredible. It was out of anyone's control, it was all automated. Then I went outside and hoped to watch it come back. I didn't have any data or anything there, I was just standing outside, waiting. It was pitch black because it was night, and then I saw the brightest light ever when the engines re-lit as the rocket came back. I was like: Oh, that's pretty close.

Margaret Laffan: Did you have to move out of the way?

Keenan Johnson: No, we were miles away and it was right where it should be. But it's still shocking because I've never seen anything like that; seeing it land. I remember it landed below the tree line a few miles away from us. When it came back, there was a sonic boom as well. And right as it went below the tree line, the sonic boom hit us because we were a few miles away. I thought it had exploded, but then I went back inside and there was a perfect video of it sitting on the landing pad. That was the best moment.

I think one of the reasons it was so good is because the previous launch had actually exploded. So this was not only the first landing but the return to flight. That was a really challenging moment, not only had we fixed what was wrong that led to that first explosion, but we then tried to do something even bigger, and it worked. I think that the whole process was many years in the making, but that was a particularly intense year for me. I had worked on that rocket quite a bit. I was in the factory when it turned on for the first time, and then all the rockets went to Texas to get tested for three months, which I also went to. I then went to Florida for two months in between being in Los Angeles where I was based. That was the most rewarding thing.

Margaret Laffan: What an incredible experience.

5. Why Impossible Aerospace?

Margaret Laffan: After SpaceX, you moved into a different venture. You became one of the founding engineers at Impossible Aerospace. What prompted your move?

Keenan Johnson:

I had been at SpaceX for quite a while and I had become really interested in climate change, specifically. SpaceX does a lot of good for climate change, but I wanted to be closer to the source of some of the problems.

I think there's a lot of different ways you can tackle climate change. Looking around, nobody was really trying focusing on aerospace. Aircraft is specifically related to climate change. Most small aircraft still burn leaded fuel, for instance, and large aircraft are big contributors to climate change. Nobody has any plans on how to solve that, electric cars are happening, that seems pretty certain; not so certain for electric aviation, especially three years ago when we started.

6. Building The Best Fully Electric Aircraft

Margaret Laffan: At Impossible Aerospace, you led the engineering team there to build the world’s best fully electric aircraft. What were the steps that you were working through to deliver such a project?

Keenan Johnson:

We started with this mission and we knew it would be hard to do, but we had a good idea that we thought would work. If you take the fuel tanks out of an aircraft and you put batteries in there, you end up with an aircraft that does not fly for very long. That's because batteries are not as great as a technology for us today. Pretty much all the major players in the space are just waiting for batteries to get better. Hopefully, that happens, that would be great. That's a super long, very technical chemistry process that needs to happen.

The other approach you could take is putting more batteries into an aircraft than anyone else, and that's what we did at Impossible Aerospace. We're going to build an aircraft that is made of batteries, which was not really an approach that anyone had looked at, mainly because it’s hard. It was exceptionally hard to think about how to do that. There's a lot of safety aspects that are difficult with batteries.

Margaret Laffan: What's your goal there? Is it to go longer, faster?

Keenan Johnson:

For the most part, it's easier to go longer or carry more stuff. I would say the most useful aircraft don't really care that much about going faster. In fact, for most commercial flights, they actually go slower to save gas. Most commercial flights can go a lot faster than they would but the airlines are trying to cut costs, so they fly slowly to save gas.

Margaret Laffan: Were you thinking commercially? Were you thinking about delivery? What were you thinking when you were approaching this? Who would use the product?

Keenan Johnson:

The goal of Impossible Aerospace is to, firstly, prove that it is possible to build a useful and fully electric aircraft. The goal of the company is to build aircraft of all skills; that's up to the size of a 787 aircraft. However, practically there were just three of us. We don't have the resources or money to build a big aircraft like that. And additionally, I think the FAA (Federal Aviation Administration) would have told us to go away. If we just started building a 787, the FAA would have never certified it. We're at the end of it. So instead, people like to talk about “minimum viable products”, what's the minimum viable product of this? For us, that was a small drone. That's what we built. It's called the “US-1”, it's out right now, it’s selling to customers. That drone can fly for up to two hours on a single charge, which is 10 times longer than most drones that fly for 20 minutes.

Margaret Laffan: What would you say is your biggest lesson from that entire experience?

Keenan Johnson: I think the biggest lesson for me personally, was how to build the organization on the people side. I think that's often just as hard, if not harder than a lot of the technical aspects. We saw large companies, for instance, with all the resources in the world like Boeing and Airbus, with infinite money to spend on this problem if they wanted to, but the real challenge is getting a creative group of people to tackle such a hard task and complete it within a reasonable timeframe.

Margaret Laffan: Many of the founders and co-founders we talked to have discussed this challenge because when you look to scalability and the infrastructure required, from the engineering to the business, to the partnership perspective, it's a very different company you start to build. Many people share the same challenge as you.

7. Challenges for Electric Aircraft Commercialization

Margaret Laffan: What do you think about the commercialization aspect? This would be more specific to aero in bringing out a product to market and so forth. We know that there's a lot of challenges around in the aviation space and many great companies are working through that right now.

Keenan Johnson:

I think the biggest challenge is going to be in safety and the certification and regulation of it. Aircraft is one of the highest regulated industries, perhaps besides the medical industry. That’s because the history of aviation is written in blood, as a lot of people say. A lot of people have died in flight. The consequences of failure are quite high. I think aircraft have basically been built in the same way since they were invented 60 years ago or so. I think society does not want to go through that period of massive failure again. Our tolerance for aircraft to fail is very low. That’s the culture, which is great, that’s ideal. The focus now is on innovating in a new way, which I think is what we're seeing in the aerospace industry - you can see that both with companies like Impossible Aerospace, or companies like Lilium GmbH which are working on flying car concepts; all of them are innovating in very different ways, but how they do that innovation is in an incredibly safe way, adding to the challenge.

8. Estimate Timeline on The Electric Aircraft Commercialization

Margaret Laffan: When you think about commercialization and taking that into account, is there any type of timeframe that you could see this being achieved by, or do you feel like we're still pretty nascent in the beginning stage?

Keenan Johnson: I think it's hard to say. What I can say is, historically, if you look at the time to develop a new aircraft from the big players, it's roughly seven years on average. But I think there may be some shortcuts. I f you could do an autonomous cargo aircraft, for instance, the safety consequences are lesser than if you're carrying people. That might be a good shortcut for some of these questions. There are some companies out there doing drone delivery right now. That's interesting, proving ground for a lot of the same concepts that we're talking about. If you can make one of those things operate safely, then that probably goes a long way towards a much larger aircraft.

9. Future Plan

Margaret Laffan: Moving on from that, what are you pursuing in terms of your current plan and your research? How are you spending your time these days?

Keenan Johnson:

Luckily, Impossible Aerospace is doing great, and I've had the opportunity to step away and take some vacation, which is nice. But I've also been refocusing on some other problems, which I think are important. I'm working on a couple of concepts right now. One is related to educational topics, regarding energy and climate change. Certainly, there needs to be a lot more information on education about those great things, especially for children, which now is really the time to start. For instance, how often do you charge your Nintendo Switch? The way you do that has real implications for the energy of the world. Those mindsets expand out and there's not really any good resource.

So I'm working on some concepts specifically around tech and robotics, of which I have a large skill set. I'm playing around with some AI projects. One of the best ways to do personalized learning is through voice. Voices are very intuitive for kids. And I think that could be a really interesting medium to communicate some of these fairly abstract and difficult concepts about energy and what that means to children in a personalized way. There's a list of engineering grand challenges. I don't know if you've read those, but one of those is for personalized learning. I think AI in voice specifically has an interesting way of tackling some of those things.

10. What Will Happen Next?

Margaret Laffan: In general, if we look into 2020, when we're thinking about aerospace, we're thinking about robotics and so forth, and you haven't been so heavily embedded in these areas. What do you predict is going to happen next year in terms of some of the biggest movements that we might see? Any Predictions?

Keenan Johnson:

I think one of the things that we will see is a budding industry of autonomous systems and robotics. I think the most interesting things are often related to robotics because they can do something physical, like self-driving cars, or self-flying cars, or delivery or surgical or whatever kinds of robotics that exist, even down to automatic sinks that wash your hands.

We don't have a good language or even concept of how to talk about building trust between those systems and the end-user, and how they integrate back into the world. For instance, if I'm a pedestrian and I see a car, I can often look into the eyes of the driver and know that the driver sees me. I know that they're not going to hit me. The self-driving car often also sees you, but we don't have a good precedent for how to communicate the same thing.

Margaret Laffan: When we think about 2020, how do you see that coming? I guess where we're going is building more trust?

Keenan Johnson:

I think we first have to develop the vocabulary to talk about these concepts. I think trust is a good way to think about it, however, there's probably a better vocabulary that’s more precise. I think that's the first step. Hopefully, the industries are going to align in some vocabulary for how we talk about the trust and reliability of these systems? And if they are not functioning correctly, how do they notify people?

Margaret Laffan: And the explainability that you bring into this as well.

Keenan Johnson:

How do they explain what's going on? I think the vocabulary will be first, and then the educational story after that. If you introduce one of these autonomous systems into a new environment, like if you had a robot that was helping people in a nursing home, for instance, how do you explain to them what the autonomous system is doing? That's going to be really important. And I think right now, we have 80% of the solutions for most things on hardware and software. Now it's more about the experience and design of these robotic systems.

Margaret Laffan: Yes, and the criticality of the function of the robot, because that's where the trust lies too. If there’s something going to knock me down or deliver the food for me, they're very, very different experiences.

Keenan Johnson:

And how do we talk about the safety of these things? It’s going to be very difficult to say this thing will function 100% of the time, all the time. Humans don't do that either. So how do we have an understandable conversation about that, in a way that the people and the regulators can use to approve of different kinds of devices?

Margaret Laffan: As we progress through 2020, as we develop this language and vocabulary, we talk more about explainability, and we can share more findings.

Keenan Johnson:

I hope and I think that 2020 will be the year we start to connect some dots. I hope the automotive industry, medical device industry, the aerospace, and other industries can find a common vocabulary and start establishing common lessons amongst them. I hope that 2020 will be the year we get to do that because I think that'll really accelerate the progress of things.

Margaret Laffan: That would be awesome. Keenan, it's been a complete pleasure talking to you today. Thank you for sharing your story and your experience. We're delighted to have you.

Keenan Johnson:

Thank you for having me.