I applied through university. The process took 4 weeks. I interviewed at Infinia ML
Interview
First, there is an hr interview to chat about your general experience. Then a tech interview to go through your resume and project experience. Next, the onsite to talk with the group members to test your domain knowledge, tech skills, and presentation ability. The last is the hm interview to ask behavior questions.
I applied online. The process took 4 weeks. I interviewed at Infinia ML (Durham, NC) in Jul 2022
Interview
The first round is an 1 hour basic ML knowledge interview. The second round is a 4 hour interview including ML case study, ML knowledge(similar to the first one), basic python coding (e.g. multiply matrix), half an hour presentation and talk to the VP.
Interview questions [1]
Question 1
Standard ML questions, not too much related to NLP and deep learning, only asked the very basics like preprocessing, BOW, word2vec. A lot of questions about OCR (I knew nothing about it, it was not in the JD either)
I applied online. The process took 5 weeks. I interviewed at Infinia ML in Jul 2021
Interview
The interview process consisted of multiple rounds including: machine learning concepts (x 2), coding skills, system design, and background. In the technical sessions, the team is working with you to find a solution to a problem. The sessions are inviting and positive!
Interview questions [2]
Question 1
What is a suitable model when you have many more features than data points? What about when you have many more data points than features?
Describe the difference between gradient descent, stochastic gradient descent, and mini-batch gradient descent. Explain the difference in computational cost. Which should you pair with a large learning rate?