
Taboola Data Scientist Interview Questions
& Process
Real candidates share what happened, how many rounds they had,
and how the experience turned out.
Based on 10 interview experiences · FREE TO READ
Candidate interview experiences
First-hand accounts from people who interviewed at Taboola.
Data Scientist
The HR person was great and easy to talk to. They have a careers page that lists out all the steps. My first interview was nice and focused on something I did at a past job. They asked a lot of ML questions. Then they told me about the company, the job, and their work. After that, I got a homework assignment with three questions: one leetcode style, one ml system design, and one about sorting algorithms. Then there was an interview to discuss the homework. The interviewer was a manager, respectful but kind of condescending. The questions were about technical stuff like why one loss function is better than another, which felt more junior level. In the last question (leetcode), they asked to add a function to delete a value from an array (the homework had me implement find and insert). Maybe there was a disconnect, I was expecting a more senior level discussion, but they wanted me to explain cross-entropy and KLD. After many interviews, this was the first time I felt completely lost and just wanted it to be over. My advice is to only spend 2-3 hours on the home assignment and just focus on studying for the interview like you normally would. The assignment isn't worth the effort.
- Explain why log-loss is better than MSE for classification tasks.
- Implement a function to delete words from a trie tree.
Data Scientist
They asked me to talk about my past work and projects. They also asked about how I like to work, if I prefer doing it alone or with a team. Then they hit me with some general deep learning questions.
- Can you tell me about a project you worked on?
- Could you describe your previous job and any projects you were involved in?
- Do you prefer working in a team or individually?
Data Scientist
Started with an HR call, then talked to the hiring manager. After that, there was a home assignment to complete. Then, an interview focused on that assignment. Next, a peer interview, and finally, an interview with the head of the department. Everything was conducted over Zoom, no in-person meetings. The conversations were pretty relaxed, not overly intense.
- Can you discuss the issues associated with sequence-based models when dealing with NLP or Time Series data?
Taboola Data Scientist Interview Questions
Quoted word for word from Taboola interview reports.
“Implement a function to delete words from a trie tree.”
Read reports →“Explain why log-loss is better than MSE for classification tasks.”
Read reports →“What's the main idea behind Boosting? Can you explain how it differs from Random Forest?”
Read reports →“Since I don't know time-series analysis, what are the difficulties in time-series analysis and how can they be solved?”
Read reports →“Can you explain how advanced neural network architectures like LSTM and Attention networks can be applied to time series modeling?”
Read report →“Can you discuss the issues associated with sequence-based models when dealing with NLP or Time Series data?”
Read report →“What are your thoughts on deep learning?”
Read report →“How did you assess the models used?”
Read report →“Could you describe the journey from an initial idea to a finished product?”
Read report →Formats, difficulty and experience
Across all 10 Taboola interview reports.