PathAI Machine Learning Engineer Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 13 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at PathAI.
PathAI
Machine Learning Engineer
Entry Average Neutral experience Accept offer 4 rounds
Interview process
Technical screen Technical screen
Interview formats
Coding Technical
It was two coding problems, pretty easy and focused on arrays, plus one debugging one. Then two ML screens, one about research and the other about systems. A new grad could handle it after about a month of prep.
Confirmed questions2 questions
- Tell me about regularization.
- Can you explain second order optimization?
PathAI
Machine Learning Engineer Intern
Intern Average Negative experience No offer 2 rounds
Interview process
Take home Recruiter call
Interview formats
Coding
Applied and was sent an hour long coding test to take a couple days later. Passed all test cases on the coding test. A few days later, I got an email saying they found a more qualified candidate before I could even talk to someone. The test is basically designed to waste students' time when they could be applying to other internships or working on projects. Good luck with this one if you're going to anything other than a school with 50,000$+ tuition. Nothing else matters to them.
Confirmed questions1 question
- Can you implement a KNN class and calculate it using both L2 and Manhattan distance?
PathAI
Machine Learning Engineer
Mid Difficult Positive experience Accept offer 3 rounds
Interview process
Take home Phone screen Onsite
Interview formats
Technical Coding Behavioral
So there were 3 interviews total. I started with an online coding challenge that had two parts: some multiple choice questions on basic classical ML and then a coding problem. The multiple choice stuff wasn't too bad, and the coding problem was like a standard LeetCode/HackerRank type thing, took a bit of effort. After that, I had a 30-40 minute phone interview that was all about deep learning, covering things like neural networks, activation layers, batch norm, initialization, and so on. The last part was a 4-hour onsite interview with the ML team and the CTO. It was broken down into 5 smaller sections: classical ML, math/probability, algorithms, deep learning, and behavioral questions. It was a really long and draining day, but overall it was a good experience.
Confirmed questions3 questions
- Questions on probability, classical machine learning, and deep learning, plus algorithms.
- Questions on probability, classical machine learning, and deep learning, plus algorithms.
- Questions on probability, classical machine learning, and deep learning, plus algorithms.
PathAI Machine Learning Engineer Interview Questions
Quoted word for word from PathAI interview reports.
“What is the purpose of using pooling in convolutional neural networks?”
Read reports →“Write a program to calculate the Jaccard Coefficient.”
Read reports →“How would you calculate the ratio of intersection over union for two sub-images taken from a given image?”
Read reports →“Could you calculate the parameters for a given CNN network based on its architecture?”
Read reports →“Regarding ResNet's skip layer, when is the right time to apply activation, and what's the reasoning behind it?”
Read report →“Can you implement a KNN class and calculate it using both L2 and Manhattan distance?”
Read report →“Can you explain why ReLU is a better choice than sigmoid?”
Read report →“Can you explain the fundamentals of deep learning?”
Read report →“Can you explain second order optimization?”
Read report →Formats, difficulty and experience
Across all 13 PathAI interview reports.