CREATORS · CAREER PROFILE

Mark Rober first reached a wide YouTube audience with a Halloween costume that made it look as if there were a hole through his torso. He was working on NASA’s Curiosity rover at the time. His later Glitterbomb and squirrel-maze videos turned that early success into a career explaining engineering through projects viewers could follow from the first attempt to the final test.
Mark Rober’s backyard squirrel videos give viewers a problem they can recognize before any engineering starts. An animal wants to get somewhere, and an obstacle is in the way. After the setup, the fun is watching how often the squirrels discover an option the person who built the course did not account for. Their choices make the design easier to understand than a diagram of the finished course would.
That approach connects Rober’s years as an engineer with his work on YouTube and at CrunchLabs. Viewers see a design and get a chance to make a prediction. Then a test shows which part of the idea needs more work. You can follow that sequence even if you arrive for the squirrels and have never taken an engineering class.
Mark Rober’s career at a glance
- Engineering: Rober says he spent nine years at NASA’s Jet Propulsion Laboratory, seven of them working on the Curiosity rover.
- YouTube: His verified channel includes series on backyard squirrels, science demonstrations and engineering builds.
- Company: He launched CrunchLabs in 2022, according to his channel biography.
- Fundraising: He co-founded the #TeamWater campaign with MrBeast, in partnership with WaterAid.
- Reported earnings: US$30 million on the Forbes 2026 Top Creators earnings list, as of June 2026. This figure measures earnings for the ranking.
Early life and engineering education
Rober grew up in Brea, California, according to a profile published by his university. He earned a bachelor’s degree in mechanical engineering at Brigham Young University in 2004 and completed a master’s at the University of Southern California while at JPL. BYU’s reporting describes him working on rover hardware, including arms and descent-stage equipment, and helping create an internal knowledge-sharing wiki. The work behind the rover was both mechanical and collaborative: a design team had to leave enough information for other teams to understand why a part had been built a certain way. BYU’s alumni profile traces those years.
From the Curiosity rover to a camera
On his verified YouTube profile, Rober describes working at JPL before spending five years in product design at Apple. He also identifies Curiosity as the project that occupied most of his time at JPL. The official channel biography is his own account of those jobs. It provides a clear timeline without implying that any one person built a rover that depended on a much larger team.

Rover engineering and a YouTube production have different stakes. The connection worth watching for is the habit of asking what has to be tested. You can see it when a piece of a build works on a bench and then fails under the conditions in which it is meant to operate. The failure gives the viewer a reason to care about the next version of the design.
What the Curiosity landing tells us about engineering
Curiosity reached Gale Crater in August 2012 after a journey of more than eight months. NASA’s mission fact sheet describes a landing that had to slow a large mobile laboratory from interplanetary arrival to a controlled touchdown. A heat shield, parachute, powered descent and a sky crane each handled a different part of that problem. At the end, the descent stage lowered the rover on cables so it could land on its wheels.
The sequence offers a practical distinction between a big invention and the parts it depends on. An effective parachute cannot tell the descent stage where it is. A stage that can hover still needs a way to place the rover safely on the surface. The landing plan had to give each component a job and define the point at which the next component would take over. NASA’s account establishes how the mission worked. Rober’s channel biography establishes his period on the team. Those are separate pieces of evidence, and neither assigns the entire landing system to one engineer.
That distinction makes a useful way to watch his later work. When a video shows a complicated machine in motion, ask what happens at the handoff between two parts. Does an object actually arrive where the next component expects it? Does the timing still work if something moves a little faster than planned? Those questions help explain why a demonstration can look effortless at the end even though the testing took time. NASA’s Mars Science Laboratory fact sheet describes the rover and landing system in more detail.
How Mark Rober makes experiments easy to follow
The backyard squirrel series is a useful place to start. As soon as viewers know where an animal wants to go, they can make a prediction about the obstacle. A replay can then show a route or a movement most people missed the first time. The point of a replay is to give the audience a closer look at the evidence. The demonstration works because it lets viewers revise their own explanation before the host supplies his.
The format also works in a large build. A roller coaster can bring questions about speed, height and the shape of a track into the same frame. Forbes cites a roller-coaster video among Rober’s projects. To see what the video is teaching, follow one of those questions from the first attempt to the final test. Notice what the team changes, and what it keeps constant so that the next result can actually be compared.
The 2020 squirrel maze as a case study
The first Backyard Squirrel Maze, published in May 2020, starts with a practical annoyance. Squirrels had been getting into the birdseed. Turning the situation into an obstacle course changes the question from how to keep the animals away to how they find a route through something built for them. The animals supply a moving target for the design because they can ignore the route a human expects them to follow.
Separate the footage of the construction from the footage of the test. During construction, the builder can explain what each part is meant to do. During the test, the animal has no reason to respect those intentions. A piece that was supposed to block one path may provide a foothold on another. The gap between the intended use and the actual use is what makes the video a useful example of iterative design. A replay gives the viewer time to spot the route that the builder missed.
There is also a limit to what the footage can tell you. A particular squirrel solving a particular obstacle is a result from that setup. It is a good reason to ask a new question, such as whether another animal would take the same route after watching. It is not enough evidence to make a general claim about how every squirrel learns. Keeping that distinction in mind makes the experiment more interesting because it leaves a concrete question to investigate.
Why the Glitterbomb projects became recognizable
The Glitterbomb series gave Rober another repeatable premise: an ordinary-looking package contains a deliberately elaborate surprise. The series became part of the channel’s identity alongside the squirrel courses. Netflix’s 2026 catalogue includes Glitter Bomb 3.0 vs. Porch Pirates as an episode of Mark Rober’s CrunchLabs. The programme also includes a pumpkin-carving competition and an episode about using sensors to explore a submerged site. That range shows how the same presentation style can move from a contraption built for a reveal to a story where the equipment helps document something. Netflix’s episode guide identifies the projects.
CrunchLabs takes the build off-screen
CrunchLabs makes engineering activity kits. For someone watching a video, an explanation can stop when the host has shown why the finished device works. A kit has to leave enough room for the person assembling it to see what happens when a piece moves, a connection changes or a planned action fails to occur.
Try separating the two jobs when looking at the company’s products. The first is to help someone get a device working. The second is to give them something to investigate once it works. A build with a clear starting point lets someone ask a better question the second time around. That is where the format most closely resembles one of Rober’s videos.
From a subscription box to classroom materials
In his TED talk My $60 million science experiment, Rober describes an effort to reach students through science instruction. TED identifies Class CrunchLabs as a free curriculum resource designed around state science standards. That is a different use of the CrunchLabs name from the subscription kits. A classroom resource has to give a teacher a way to lead a lesson and make the learning goal visible to a group of students.
One way to evaluate such a lesson is to ask what students can explain afterward. Getting a build to move is satisfying, but recognising which change made it move is a different achievement. A teacher can ask one group to keep the build as it is while another changes one variable, then compare the outcomes. This turns the lively part of a demonstration into a question students can answer using what they saw.
For readers working through Rober’s career, the rover explanation, the squirrel maze and the TED talk make a useful sequence. The first shows the responsibilities of a team working under fixed mission constraints. The second shows how an unknown result can make a video compelling. The third shows how an explanation can be designed to leave the next test in the learner’s hands.
TeamWater and the work beyond the fundraiser
WaterAid records Rober and Jimmy Donaldson, known as MrBeast, as the founders of #TeamWater. It says the campaign reached its US$40 million goal in August 2025. The fundraiser paired creators’ public reach with a charity that could describe how money would be used. The amount raised measures the campaign. WaterAid’s project updates are the place to follow what that funding delivers over time.
For the other side of that collaboration, read Abroviaon’s MrBeast profile.
The Halloween costume that turned into a business
Rober’s early costume used two iPads on a FaceTime call. Each screen showed what the camera on the other side could see, creating the illusion that viewers were looking through his body. He went on to sell digital costume designs and eventually sold the business to Morphsuits in 2013. It was an early example of a pattern that would show up again at CrunchLabs: a video gives people a reason to pay attention, and a product gives them a way to try the idea themselves. BYU Magazine’s account of Rober’s early videos and business.