Optum ML Engineer interview typically runs 1 round: intro and technical interview. Based on one report, it took about 1 call and felt highly rigid with frequent interruptions.
$135K
Avg. Base Comp
$167K
Avg. Total Comp
5 rounds
Typical Rounds
1-2 weeks
Process Length
Our candidates report that Optum’s ML Engineer interviews can feel less like a collaborative technical discussion and more like a test of whether you can stay tightly aligned to the interviewer’s preferred structure. In the experience we saw, the interviewer repeatedly redirected the candidate to answer in a problem statement → implementation → impact format, and kept pushing for a full end-to-end explanation without letting the candidate finish. That pattern matters: Optum seems to value candidates who can narrate work in a very controlled, business-first way, not just those who know the model details.
A recurring theme is the emphasis on health-domain specificity. Even when the candidate said the domain was outside their background, the interviewer continued probing there rather than pivoting to transferable ML depth. That suggests the bar is not simply “can you build ML systems,” but “can you defend your choices in a healthcare context and keep the conversation anchored to the use case.” We’ve seen that this can become a make-or-break point for strong generalists who haven’t worked in the exact domain.
The other signal is interpersonal: the interviewer’s interruptions were frequent enough that the candidate felt they could not complete a thought. For us, that means preparation here is not just about content, but about being able to stay concise, reset quickly, and keep your explanation tightly sequenced even when the conversation is being steered hard. Candidates who do best at Optum are likely the ones who can make their ML work sound operational, domain-aware, and immediately relevant under pressure.
Synthetized from 1 candidates reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Optum process.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Optum
Write a function `sorting` from scratch to sort a list of strings in ascending alphabetical order
| Question | |
|---|---|
| String Mapping | |
| String Palindromes | |
| Impossibly Iterative Fibonacci | |
| Your Strengths and Weaknesses | |
| Prime to N | |
| Hurdles In Data Projects | |
| P-value to a Layman | |
| Bagging vs Boosting | |
| The Brackets Problem | |
| Random Forest Explanation | |
| Find Duplicate Numbers in a List | |
| Valid Anagram | |
| Assumptions of Linear Regression | |
| Drawing Balls From Bin | |
| Bias vs. Variance Tradeoff | |
| Most Repetition | |
| Data Preparation for Imbalanced Data | |
| Multicollinearity in Regression | |
| Prime Numbers Identification | |
| Skewed Pricing | |
| Support Vector Machines vs Deep Learning Models | |
| Type I and II Errors | |
| Implementing the Fibonacci Sequence in Three Different Methods | |
| Why Do You Want to Work With Us | |
| Swap Variables | |
| Decision Tree Evaluation | |
| Combinational Dice Rolls | |
| Stakeholder Communication | |
| Risk Assessment Model |
Synthesized from candidate reports. Individual experiences may vary.
The interview starts with a short introductory conversation to set context for the role and the candidate’s background. In the reported experience, this was only a brief opener before moving directly into a project discussion.
The interviewer asks the candidate to walk through a recent AI/ML project in a structured way, specifically as problem statement, implementation, and impact. The candidate is expected to explain the work clearly and concisely, but the conversation may be tightly controlled and interrupted frequently.
After the project overview, the interviewer pushes into the end-to-end ML pipeline and expects a complete explanation rather than partial answers. In the shared experience, the interviewer repeatedly interrupted mid-answer and pressed for full technical detail across each phase of the workflow.
The interviewer may shift into domain-specific questions tied to healthcare or a related health context. The candidate in the experience was asked questions outside their background, suggesting the discussion can become highly specific to the interviewer’s chosen domain.
The process ended abruptly in the reported case, with no offer extended after the technical conversation broke down. Based on the experience, there was no indication of additional rounds after this interview.