Coupa Software Inc logo

Coupa Software Inc Data Scientist Interview Questions
& Process

Real candidates share what happened, how many rounds they had,
and how the experience turned out.

Based on 5 interview experiences · FREE TO READ

4.2 Rounds average
Average Typical difficulty
80% Positive experience

Candidate interview experiences

First-hand accounts from people who interviewed at Coupa Software Inc.

Showing 3 of 5
Coupa Software Inc logo
Coupa Software Inc

Data Scientist

Analytics · nearly a year ago

Mid Average Positive experience No offer 1 round
Interview process
Recruiter call Phone screen
Interview formats
Behavioral

The interview process was pretty straightforward. I had a call with an SDE 2 who was really nice and helped me relax. They sent over a link for the meeting, and they joined right on time.

Confirmed questions2 questions
  • Could you describe a project you particularly enjoyed working on?
  • Tell me about a project where you learned the most.
Coupa Software Inc logo
Coupa Software Inc

Data Scientist

Analytics · more than a year ago

Mid Easy Negative experience No offer 6 rounds
Interview process
Recruiter call Phone screen Technical screen Onsite Panel Background check Offer
Interview formats
Behavioral Technical Coding System Design Case Presentation Group

Applied online and the first interview was with a recruiter who basically just repeated the job description and couldn't answer any questions about the role, company, or team. After passing that, my resume went to a higher-up who asked detailed questions about my resume, which was fine. The next round was rough. Scheduled for 30 mins, but the interviewer extended it to an hour and immediately started asking questions without introductions. They hit me with tons of Computer Science and software engineering questions, like how to make a multithreaded app in Python and asking for descriptions of all major data structures. This went on for over 2 hours, making me miss other appointments. The interviewer seemed to want to leave before I could ask questions, despite me answering everything. When I finally asked about his role and how he keeps up with Deep Learning & NLP advancements (which were in the job description), he claimed he hadn't learned anything new since the early 2000s and that this role just uses .fit() and .predict(). He was the lead data scientist on another team and had no prior DS/research roles. He also told me I wasn't good enough for Coupa, even though I was moving to the next stage. He then asked about Kubernetes, and when I explained my experience with Minikube and Azure's Kubernetes for scaling an API, he questioned how I knew about it and didn't seem to believe I had experience. The next round had 2 data scientists who were a bit nicer but still not great. By this point, I had another offer and wasn't as invested, but I went in with an open mind. One interviewer asked basic deep learning questions (RNN vs Transformer) but no deep dives into model selection, embeddings, or optimizers. He then argued with me about a project's impact at my current company, acting like he knew more than I did. Then came brain teasers, which I didn't bother with due to the interview vibe and his lack of knowledge. This interview was at 6:00 am PST because he was in another country, and they mentioned this early/late meeting schedule is common (another major red flag). The next interview, 1.5 hours later, was mediocre, with an uninterested interviewer. They had a weird exercise where I had to create a database and write a SQL query without knowing the database purpose or query goal beforehand. They didn't ask for anything in writing, just the exact syntax explanation, which was unusual compared to other tech interviews. This was supposed to be one of the final rounds, with a total of 6 rounds plus a hiring manager interview mentioned. I was rejected two weeks later. As a final note, when I asked about AI/ML learning initiatives, they said they haven't set any up despite the org being around for 5+ years.

Confirmed questions7 questions
  • Can you explain the difference between an RNN and a transformer?
  • What are embeddings?
  • Why do we use optimizers instead of vanilla gradient descent?
Coupa Software Inc logo
Coupa Software Inc

Data Scientist

Analytics · more than a year ago

Entry Difficult Positive experience Accept offer 2 rounds
Interview process
Recruiter call Technical screen
Interview formats
Technical Behavioral

The interview was very detailed and technical. It was friendly too. I was nervous at first but they made me feel comfortable, and then I was able to answer their questions normally.

Confirmed questions2 questions
  • What were your internship experiences like?
  • Can you describe the projects you've worked on?

Coupa Software Inc Data Scientist Interview Questions

Quoted word for word from Coupa Software Inc interview reports.

How do you calculate the number of trainable parameters in a CNN?

Read reports

Why do we use optimizers instead of vanilla gradient descent?

Read reports

Can you explain the difference between an RNN and a transformer?

Read reports

Describe every major data structure, such as Trees, Tries, Hashmap, Heaps, Stack, List, LinkedList, Hashset, etc.

Read report

Can you explain how to create a multithreaded application in Python?

Read report

Can you create a database and write a SQL query for it?

Read report

Formats, difficulty and experience

Across all 5 Coupa Software Inc interview reports.

Interview formats

Behavioral 29.4%
Technical 29.4%
Coding 17.6%
System Design 5.9%
Case 5.9%

Interview difficulty

Easy 20%
Average 40%
Difficult 40%

Candidate experience

Positive 80%
Negative 20%