Over the past 7 years, Mallory Lefland has worked in various capacities at NASA’s Jet Propulsion Lab (JPL) — most recently on the MARS 2020 project that recently celebrated the successful landing of the Perseverance Rover and Ingenuity Helicopter on the red planet. Her main jobs as an aerospace and systems engineer involved conducting multiple tests to find potential flaws and make improvements on the project. The Press recently sat down with Lefland over Zoom to talk about her career, her ambitions and how it felt to be part of NASA’s latest Mars mission.
I saw you talk about this a little in the video you did for NASA, while working with Curiosity — but what got you interested in aerospace engineering in the first place?
I’m not really sure how I settled on aerospace engineering. Looking back, it seems pretty random — but I think it had to do with my love of airplanes, my love of flying. I also thought that aerospace engineering was really specific, and I wouldn’t have to make a lot of career decisions after that point — which sounded nice. But as you get into any field, you realize how vast and varied every kind of job is. You really still need to pinpoint exactly what it is you want to do. So I don’t know, I wish I had a great story for how I got here but it all feels pretty random and fortuitous, to be honest.
I mean, I think there’s a lot of people who can pinpoint it. I have friends who watched Star Trek and saw a specific thing and said “I want to do that”. And then they have their sights set on NASA. And I definitely know that there were definitely a lot of people I was in college with who wanted to be pilots, or wanted to work for NASA and kind of had that goal their whole life. I just think there’s an equal number of us who just sort of fell into it.
I understand that becoming a professional engineer is a four year program, plus four years of shadowing another professional engineer, as an engineering intern. Is that how it went for you?
So, I’m not a professional engineer. There are some fields where you need to be a professional engineer, I think in civil engineering, in order to sign specific documents for buildings and things like that. For aerospace engineers, I don’t know a single aerospace engineer who goes through an FE (Fundamentals of Engineering) or PE (Principles and Practice of Engineering) test*. That’s not like a requirement for any aerospace job I’ve ever seen.
So you’re saying you did a lot of undergrad internships, completed a bachelor’s degree at Georgia Tech, and after that you were qualified?
Yeah. I did a bachelor’s degree in aerospace at Georgia Tech, and then was hired at JPL to be a systems engineer right out of college.
How would you describe your role in building the new EDL (Entry, Descent & Landing) technologies used on Perseverance? I remember one of the big things that everyone seemed to be talking about was its ability to autonomously assess landing hazards. How would you describe your role in building those into the EDL technologies to non-engineers?
So I didn’t work on the TRN (Terrain-Relative Navigation) development. There was a team of people who were specifically working on developing the hardware and software that did that. My job mainly pertained to making sure that the overall EDL flight system could land properly and was safe under a variety of different software and hardware behaviors.
Once the TRN system was integrated into the March 2020 EDL system, my job included running a lot of landing tests to make sure we could land safely. And then I ran many, many — dozens — of off-nominal testing, where I would create faults in the TRN system to make sure that EDL could still be successful and safe — regardless of the kind of faults we’ve seen. Anytime you add a piece of hardware, or some amount of software that’s new into the system, you have to create tests that target that specific area, of hardware or software, to make sure that faults in those areas don’t ripple through the rest of the system.
So was this testing in simulations. Or was it prototypes of the rover being tested in similar conditions to Mars?
We have a lot of different test environments starting out from, like, 100% simulations that I could run on my computer; then we have testbed environments — where we have copies of all of the hardware and the guts that are inside the rover and inside the cruise and descent stage, and you’re able to test landings using copies of all of the hardware; all the way up to when we have the fully-assembled spacecraft — the real spacecraft that will go to Mars.
We have a program that’s called ATLO** (Assembly, Test and Launch Operations), which is where we run tests on the actual spacecraft before and after we send it to Florida, and before it launches. So, in that case, we could run a landing test on the real spacecraft. If you were to look at the spacecraft when it was happening, it’d be really hard to tell that it’s going through landing because we’re obviously not, you know, separating devices and it’s not flying, but the computer is running through the entire landing sequence. We’ve just kind of stopped it from doing most of the hardware reconfigurations. I did tests on all of those different kinds of test beds.
So it’s actually the one that’s going to be going into space?
Yeah, so ATLO is when you’re actually building the spacecraft. You start building it piece by piece — and along the way, as you’re putting it together, you test it in a variety of configurations. But you are testing the hardware and software that…will be going to Mars.
So you’re testing it in a variety of conditions, to see if any flaws can be found.
We test for a lot of different reasons.
You first put a test campaign together to verify all the requirements you wrote on the system. So when we’re designing a system, we start at the requirement level that says: here are the thousands of things that the spacecraft needs to do. You then have to either analyze or test and make sure that your final design satisfies all of those requirements. But requirements don’t tell the full story — because when you have some system that’s incredibly complicated, when you kind of add up all of the smaller pieces, you oftentimes get a system that doesn’t behave exactly how you wanted it to behave. Because when you add up all of the little pieces, there’s inherent, emergent behaviors that exist that you couldn’t have predicted during the requirement phase.
We do a lot of large scale validation and system testing, just to put the spacecraft through its paces, and then we do a bunch of off-nominal testing — which is testing where you are trying to understand the limits of the system, how robust it is to any sort of external or internal fault. And if you find during that testing, that there are faults that are unrecoverable, and you believe that they are worth preventing then you would go in and try to make the case for fixing it. But sometimes you have to live with it and say, “we could get a really bad day, we could get this specific kind of fault at this very, very specific time, and we wouldn’t be able to land successfully.” But here’s so many of those little things that can happen and you can’t prevent them all.
So as systems engineers, a big part of our job is going through a risk assessment of everything that can kill us. And understanding the probability that those things could happen, and what’s the consequence of them and trying to make the case of which ones we fix and which ones we don’t.
So something could happen, but it’s below the threshold of probability for things we can reasonably spend resources on.
I mean, in a perfect world with unlimited resources and time and people and money, you would fix everything. But you’d keep finding things that could kill you, right? You’d keep finding little things that the spacecraft wouldn’t be able to survive. But with unlimited resources, you’d continue to fix all these things.
We obviously don’t have that. No one has that. Plus, any time you go into the system to make any kind of change, you’re inherently creating some sort of ripple or butterfly effect where you’re fixing one thing, and you’re gonna cause a problem somewhere else. So it’s hard to go in and spot fix these problems without being completely sure you’re not causing problems in other places.
Were you watching the live landing the same way that it was shown to the public?
I was on the console during landing. We all had specific jobs in terms of what we were looking for, and what our call outs were and what our specific roles were on the day of landing. So I was watching the telemetry flow down, but the public’s watching people both in the cruise Mission Support area, as well as [in] other rooms at JPL. They’re watching people who are on console monitoring the telemetry. The people who are on console — we’re not watching the feed, we’re watching the spacecraft.
How did you feel while you were watching your console once the descent started?
I mean it’s strange. It’s a lot of emotions that are sort of rolled into one. I felt really nervous and scared, [and] obviously a big sense of relief at the end when we did land. The hardest part for me was the data that we were seeing during landing looks almost exactly the same to the data I see in all of the tests I run. I had seen that same telemetry hundreds and hundreds of times in the last few years — so I had a really hard time wrapping my head around that it was real. I would just look at it and think ‘Oh, of course, this is happening. This happens in every simulation and every test I run.’
And I had to kind of switch my brain to say, like, no, this is for real, this is actually happening on the spacecraft. When I run a test and I see a message about cruise stage separation, it obviously means in my test that the cruise stage is separating, but nothing really is happening. So when I see that same kind of message, on the day of landing, I had to tell myself, “Oh, that means we just actually, you know, initiated a bunch of pyrotechnic devices.”
And now they’re like, “we separated a piece of hardware”. And that piece of hardware just flew off the spacecraft. Like that happened for real, even though it looks the same as it did in every simulation. So I had a bit of a hard time realizing that it was real.
Still doesn’t feel real, to be honest.
So, you also worked on the Curiosity rover. What’s the difference between Curiosity’s mission and Perseverance’s goals?
The Curiosity rover launched in 2012, and it brought with it a suite of instruments. It was looking for signs of what could have once been a habitable environment for life. What they tried to do on that mission was to take a bunch of science tools and experiments that you would have in a lab on Earth, shrink them down and put them in a rover, bring them to Mars and try to analyze rocks there. What they realized is there’s so much more you can do in a lab on Earth than you can do on Mars.
One of the main goals of the Perseverance Rover is bringing a system with it that can core into rocks and put rocks into tubes, hermetically seal them and then leave them on the Martian floor for another future mission to come pick them up and bring them back to Earth. And the idea there is that if we ever want to determine if there’s life on Mars, or if there was life on Mars, the easiest place to do that would be in a lab on Earth. So we have to bring the samples back.
Are any of these missions for bringing back samples planned yet?
There’s currently a Mars sample return campaign that’s in the works — it’s in the early stages of development, intended to launch in 2026. And it has a lot of partners. It’s a mix of NASA and ESA (European Union Space Agency) — a few different NASA sites and JPL have a big portion of that mission.
What was your role in the Curiosity rover building?
I didn’t have any role in the development of Curiosity. I was in college when it landed, starting my final semester at Georgia Tech. I started working on the engineering operations team about six months after it landed on the surface of Mars.
I had a few roles during the year and a half that I worked on it. I helped analyze the data that came down from the spacecraft every day, I helped build the engineering plans on what we would do on the spacecraft. I developed activities and plans for how to keep a variety of hardware safe and checked out — specifically, the hardware that was redundant that we weren’t using all the time.
And then kind of on the side, my major project was — there was a set of flight software capability involving controlling the thermal system that would make [it] much more efficient, but it hadn’t been tested yet and it hadn’t been approved for use on Mars. So I spent probably about a year putting that software through a full suite of testing to verify that it could be used, and then I built the first plans for it to execute on Mars, and now they use it. I think they still use it every day.
That thermal system — is that for keeping the electrical components at the right temperature, because Mars is very cold?
Mars is like a pretty extreme desert. It gets very cold and very hot. So depending on what device you’re talking about, you need to keep it within some thermal range. This takes a lot of CPU (Central Processing Unit) power — it takes a lot of energy to make sure that you’re heating things properly. There’s different algorithms that you can use and different monitoring systems to keep all of the hardware in the right thermal zones, and the one I was working on was just like a different type of algorithm, it just hadn’t been verified yet for flight.
So what have been the most difficult obstacles to overcome in your career, the points where you almost considered going into another field?
The Georgia Tech aerospace program was really hard. They do that on purpose — weed people out and make sure that if you are graduating from the school, they’re sure of your skills when you go into the workplace. There were definitely points in time where I considered changing majors and there are certain classes where you just think that you’re never going to get this. There were certain semesters where I was studying for one class, I kid you not, 25 hours a week. But sometimes you just have to do that.
So the degree itself was an endurance challenge. What about your NASA work?
Probably the hardest part about working on Perseverance, particularly? Probably two things:
One is that you’re asked to give a lot of yourself to the project, to make sure it can land safely. The schedule can be grueling for years at a time. There are certain things where you have to plan your life around the spacecraft instead of the other way around. It’s one of those things where everyone on the project is doing it — but when I try to tell my friends about it, who don’t work at JPL, it’s hard to explain. It’s hard to explain how much you have to give up personally, in order to make sure that the mission is safe. So that can be a struggle.
And then I think for me, the harder part is you become very emotionally connected to the spacecraft. It kind of takes on a life of its own, because you’re giving up so much of your time to it. Sometimes it’s hard to have that separation in place between work and life, and you get really involved. One of the things I really struggled with was, if the mission wasn’t successful, what kind of emotional damage would I have to work through? I’d put so much of my life into this project, I was so emotionally connected to it, that the idea of it not succeeding would be incredibly difficult for me, both professionally and personally. That’s just one of the things that’s hard.
You work really hard on a mission, you get very connected to both the people and the spacecraft itself, and then you kind of have to work through stuff, if it’s not successful, or even when you stop working on it. Like right now, my EDL job has sort of ended and it’s strange.
You get very connected to both the team and the spacecraft itself while doing your job, to the point where you can worry about the effects on your mental wellbeing.
I worked on the mission itself for about seven years, [and] I worked with the EDL team for the last four or five years. You’re so focused on this specific 45 minutes and you’re just striving every day to make sure that it’s safe.
We have a lot of people who run a lot of different tests for years. It might be just you and the spacecraft for eight hours, you’re just sitting in there running a test. I probably racked up like 2300 hours with the spacecraft, running tests on the actual hardware or running simulations on my computer. It becomes much more than hardware and software; you become more connected to it just because you’re spending so much time with it. And with everyone around you, it’s sort of the same way.
I’ve called people at 3 a.m. many, many times. I’ve been woken up at 3 a.m. many times because something weird is happening and they need my help. I’ve called people when they’re on vacation. I’ve FaceTimed people at 2 a.m. You form this bond with everyone who’s going through that for years, right? Then when you stop working on it, or if it was, God forbid, not successful, there’s a real part of you that will struggle with that.
Do you have any specific role models or mentors you want to acknowledge?
No one in particular. There’s just certain people who I worked with both on Curiosity and on Mars 2020 (Perseverance’s mission). And also just people I see at JPL. No one specifically. There’s people you see, and you really admire their leadership skills or their technical skills, or how they’ve managed to combine the two of them into truly powerful roles.
Perseverance has been on Mars for 49 total days since it touched down at Octavia E. Butler Landing site in Jezero crater in February. On April 3, Perseverance deployed the Ingenuity helicopter, which will be the first aircraft to fly on the red planet. As for Lefland, while she isn’t sure where her career will take her next, she hopes to expand her skillset with an eye towards the future.
“It’s sort of hard to control your fate a bit,” she said. “But I think for me, what I’m just focusing on right now is trying to build like different types of technical skills. In the next jobs I take, to build more of a technical foundation for future roles.”
*These are the two tests you have to pass to be licensed as a professional engineer.
**ATLO is a phase in the process of making sure a spacecraft is ready for spaceflight, not a software program.