NDA for coding questions, but a lot of brain teasers during phone interviews.
Ai Software Engineer Interview Questions
6,013 ai software engineer interview questions shared by candidates
What is the difference between BERT and GPT architectures?
Standard 'Introduce yourself' type of questions; Background domain-knowledge questions (related to autonomous driving); small case studies on AV system deploiements
Leetcode style questions are perhaps the most important for any software role in Meta
Coding, ML system design. They ask to design a recommender system for a service like Google Maps.
How would you feel working for Meta from an ethical standpoint?
Name a time where you had to solve an unclear problem halfway through a project.
Do you have a work visa for the United States?
1. Moving given an array. What if instead of array it's a stream of numbers? 2. Write an iterator class for binary tree
Since the interview was purely based on your current work experience and follow up from that so my interview questions were based on the same. Below are my questions(may help you estimate yours): There is no intro in the call, straight ques-ans round. 1)Let's begin with your experience with AWS. Can you tell me about a project where you utilised AWS services and which services you found the most important for your project's success. 2) Shifting focus to your NLP experience, could you elaborate on how you have applied NLP techniques in your project, particularly any challenges you faced in language processing and how you overcame them. 3) Given your experience in Python, can you tell us how you have used Python in conjunction with NLP libraries to preprocess and analyse text data on your project. 4) Regarding your work with BERT and other transformer model, could you specify specific incidence where you applied BERT to a problem and how you fine tuned the model to serve your specific use case? 5) How did you specifically fine tuned BERT to your use case? How did you handle the training doc and what were your outcomes after applying BERT to your text data. 6)Elaborate on fine tuning process, how did you decide the hyperparameters and what kind of improvements did you see after fine tuning compare to before. 7)Can you tell me about a time where you had to optimise the model performance post deployment, perhaps in cases to data distribution or user feedback, how did you approach the issue and ensure the model remained accurate and relevant?
Viewing 5631 - 5640 interview questions