
Onfido Machine Learning Engineer Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 7 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at Onfido.
Machine Learning Engineer
I was contacted a week after submitting my application. An internal recruiter scheduled the initial 30-minute phone call, but they were half an hour late. During this call, they inquired about my salary expectations, my reasons for wanting to join the company, and detailed the employee benefits. Following this, a similar call was arranged with a more technically-oriented individual. This conversation mirrored the first, outlining the subsequent steps: a technical test would be administered, and I'd be contacted the following week to arrange a date for receiving the 24-hour take-home assignment. Despite waiting for the email, it never arrived. Subsequently, the original recruiter set up another phone conversation. We confirmed a date and time, but the call never happened! I haven't received any further communication since.
- Why do you want to work for us?
Machine Learning Engineer
The whole thing took about 3-4 weeks. Started with a nice recruiter intro call. Then a call with the head of the team, talking about the role and company at a high level and my experience. After that, I got a take-home test. Then, a call with two engineers where we discussed the test, my background, and MLOPS. The last part was a 1.5-hour system design interview. The recruiter kept me updated with feedback after each stage. In the end, they went with someone who had more engineering experience.
- Can you design Onfido from scratch?
- What are the methods for checking model performance in production?
- How would you approach the Googley eyes assignment?
Machine Learning Engineer
The phone interview happened at a time we'd set up, lasting about 45 minutes. Not much chit-chat, the interviewer jumped straight into technical questions. When I stumbled, he'd tell me what he was looking for. It was a decent interview, but I think I messed up on some pretty basic stuff. Got an email a few days later saying they weren't moving forward with my application.
- Can you explain general ML concepts like regression and random forests?
- What exactly is a neural network and what are its key characteristics?
- In which scenarios are neural networks a good fit, and when should they be avoided?
Onfido Machine Learning Engineer Interview Questions
Quoted word for word from Onfido interview reports.
“What exactly is a neural network and what are its key characteristics?”
Read reports →“Can you explain what a smart pointer in C++ is?”
Read reports →“In which scenarios are neural networks a good fit, and when should they be avoided?”
Read reports →“Could you walk me through the backpropagation algorithm?”
Read reports →“What strategies could be employed for an image classification problem when you only have about 20 images to work with?”
Read report →“What are the methods for checking model performance in production?”
Read report →“Can you explain general ML concepts like regression and random forests?”
Read report →“Can you design Onfido from scratch?”
Read report →“How would you approach the Googley eyes assignment?”
Read report →Formats, difficulty and experience
Across all 7 Onfido interview reports.