
Unity Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 29 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at Unity.
Data Scientist
Started with a phone screen, then an onsite with 5 interviews. After that, they wanted to do a reference check, which I agreed to. They eventually gave me an exploding offer. I asked for more time to decide since I was interviewing elsewhere, but the recruiter was very pushy. I managed to get a few extra days, but then they rescinded the offer, citing my 'unprofessional-ism'. I don't get how asking for time to consider was unprofessional, but I guess I dodged a bullet. The interviewers were nice, but the recruiter's behavior was a red flag.
- Can you discuss your past experience?
- Perform a coding exercise.
- Meet with the Hiring Manager.
Data Scientist
The interview process was pretty standard. First, a 30-minute video screen with the recruiter, then shortly after, a 45-minute video interview with the hiring manager. They asked some technical and behavioral questions, which gave me a good sense of the role's expectations. Next was a take-home assessment that wasn't too difficult. After that, I had the virtual on-site interview loop. This included a 30-minute interview with both a senior product manager and a senior engineering manager simultaneously, and then two separate 45-minute interviews with data scientists. The recruiter was great and always ready to help with questions. I received feedback after my interviews, which showed me how transparent the company is.
- Can you discuss technical details about projects you've undertaken and the architecture of the models used?
- Describe your thought process for constructing a hypothetical model, including the technical aspects of the projects you've worked on and model architecture.
Data Scientist
The interview process started with a recruiter call, a typical 20-30 minute chat about the job, why Unity, and my background. Then came an initial interview, about an hour long, covering conceptual DS/ML topics and some SQL coding. The final stage involved a series of 4-5 webcam interviews on Coderpad. These included one with two SQL questions, another with a Python coding task (a simplified K-Fold Cross Validation), a meeting with a manager to discuss potential projects, and a session working through a math/ML problem, like normal distribution. I was really impressed by everyone involved and chose Unity over another offer because of this experience.
- Can you code a simplified k-fold cross validation in python?
Unity Data Scientist Interview Questions
Quoted word for word from Unity interview reports.
“Can you simulate a fair coin outcome using a biased coin?”
Read reports →“Can you code a simplified k-fold cross validation in python?”
Read reports →“How would you traverse a tree to check if it satisfies a certain property?”
Read reports →“For the take-home challenge, what loss function did you select for the random forest model?”
Read reports →“Could you please solve the task using the provided dataset?”
Read report →“Can you explain the basic principles of reinforcement learning and how it can be applied to real-time ad bidding scenarios?”
Read report →“Create a Jupyter notebook with an ML model to address a typical business challenge for the team.”
Read report →“What are your thoughts on logistic regression?”
Read report →“What are the usual rounds for tech in the bay?”
Read report →Formats, difficulty and experience
Across all 29 Unity interview reports.