
Mujin Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 45 interview experiences · FREE TO READ
Which role are you interviewing for?
14 roles · 45 reportsCandidate interview experiences
First-hand accounts from people who interviewed at Mujin.
Robotics Software Engineer
This was a terrible experience that wasted 3 months of my time. They called me onsite for the final round, but then changed my position without informing me and offered a role I didn't even apply for. It felt planned from the start. We should stop lying and be ashamed. This isn't a game, and people's efforts shouldn't be played with. It started with a technical MCQ on Python, C++, and linear algebra. Then came a live coding challenge. After that, a 48-hour challenge to build a legacy system in Docker and create a REST API for it. The final round was onsite, with 3 interviews, a factory tour, and dinner.
- How would you implement an LRU cache in C++?
- What are your ideas for gripper design for robotic arms?
- We misled you about the position and are now offering a different role. Will you accept?
Software Integration Engineer
The interview process had three rounds. The first was a coding round, pretty easy with a few MCQs and two coding problems. The second was a virtual interview covering the basics of robotics and path planning algorithms. The third was a 48-hour challenge, which was tough because of dependencies in a legacy container, but I managed it in about 36-40 hours. After that, HR told me I wasn't moving forward, and they took 18 days to get back to me after the challenge.
- Can you explain what a Jacobian is?
- What is RRT?
- Can you describe CBF?
Computer Vision Engineer
So there were three rounds. It started with HR, then a technical interview, and then a final chat with the CTO. The HR part was pretty basic, mainly about salary expectations and whether I needed visa sponsorship. After that, I had a technical interview with the Director of the Computer Vision Team. This focused on the core concepts of Geometric Computer Vision, like the Essential, Fundamental, and Homography matrices, and Camera Intrinsics. Interestingly, they didn't ask any questions about ML-based Computer Vision, though the Director mentioned they're open to using ML methods. It seems their focus is more on traditional robotics and factory automation, similar to companies like FANUC or Cognex.
- Can you explain the differences between fundamental, essential, and homography matrices?
- What are the constituents of a camera intrinsics matrix and how do you perform lens distortion correction?
- What units are used for the elements within the camera intrinsics matrix, and what is the impact of cropping an image on these intrinsics?
Product Designer
It was a pretty standard process and the HR person kept me in the loop the whole time. First, I had a chat with HR, then with the lead product designer. After that, I presented a case study to some other designers, and finally, I interviewed with the lead front-end engineer. Weirdly, the engineers asked the toughest questions. I didn't make it to the last round, but HR let me know, which was cool.
- Be ready to explain your design process.
- They will also inquire about projects that were unsuccessful.
- How would you adapt the same base UI for a tablet compared to a desktop?
Mujin Interview Questions
Quoted word for word from Mujin interview reports.
“What units are used for focal length and the principal point in a camera matrix?”
Read reports →“What are the key differences between a SIFT descriptor and a SURF descriptor?”
Read reports →“What units are used for the elements within the camera intrinsics matrix, and what is the impact of cropping an image on these intrinsics?”
Read reports →“What constitutes a keypoint, and what are the uses of keypoints in computer vision?”
Read reports →“What is a homography, and how is it utilized in computer vision?”
Read report →“What is a convolutional neural network, and how is it applied within computer vision?”
Read report →“What are the constituents of a camera intrinsics matrix and how do you perform lens distortion correction?”
Read report →“How would you compute the transformation matrix linking the camera and the end-effector?”
Read report →“What is a convolution, and how does it apply to computer vision tasks?”
Read report →Formats, difficulty and experience
Across all 45 Mujin interview reports.