Credit Karma Data Scientist candidates describe a recruiter and hiring-manager screen, a coding and ML round, then technical and case-style onsites centered on recommendation systems, ML fundamentals and ML system design.
$207K
Avg. Base Comp
$350K
Avg. Total Comp
Not reported
Typical Rounds
2-3 months
Process Length
Across four candidate reports, the Credit Karma Data Scientist interview leans heavily on recommendation systems. Three candidates describe case studies or deep technical discussion on ranking and recommendation. A fourth was asked in a hiring-manager call how they would evaluate recommenders and select target populations.
The early stages look fairly consistent. Candidates describe a recruiter screen and a hiring-manager conversation. Some then report a coding round: one mentions a LeetCode-style dynamic programming problem, another a number-of-islands problem paired with SQL, and another LeetCode-medium coding alongside ML basics.
The virtual onsite is where reports diverge. One candidate had three 45-minute technical interviews covering ML Ops, ML basics and a case study. Another described ML fundamentals, MLOps system design and statistics in one day. A third described two recommendation-focused rounds. Two candidates also describe a functional or behavioral onsite that turned technical. In one case a director ran a recommendation-system case, and in another the final hiring-manager round presented ranking and recommendation case studies with probing follow-ups.
To prepare, practice explaining recommender design end to end: two-tower models, embedding choices, loss design, kNN versus ANN, and how each choice ties to a business metric. Also rehearse binary-classification cases and MLOps deployment questions. Candidates report that timelines ran about two to three months, with slow scheduling and unclear next steps, so ask recruiters early what each round covers. All four reports ended without an offer.
Synthesized from 4 candidate 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 Credit Karma process.
I started with a recruiter call that was pretty standard — mostly going through my background and relevant experience. That seemed to go fine, so I moved on to a follow-up conversation with the hiring manager. Again, it was focused on my background and qualifications, but this time I could tell something was off. Toward the end of the call, the hiring manager was pretty direct: they didn't think my background aligned with what they were looking for in the role. They mentioned some specifics around recommendation systems and how I'd approach evaluation and target population selection, but it felt more like they were testing fit than genuinely exploring whether I could grow into the role. The rejection came fairly quickly after that. It wasn't a grueling multi-stage gauntlet, but it was clear early on that the bar for background fit was pretty high, and I didn't clear it.
Prep tip from this candidate
If you interview here, be ready to discuss recommendation systems deeply — specifically how you'd evaluate them and design target population selection strategies. Make sure your background explicitly demonstrates hands-on work in this domain, or be prepared to articulate how your adjacent experience translates directly.
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 Credit Karma
How would you set up this test?
| Question | |
|---|---|
| The Brackets Problem | |
| Type-ahead Search | |
| Testing Price Increase | |
| Offer Matching API Design | |
| A/B Testing a Checkout Button Change | |
| Walking Robot | |
| Check Matching Parentheses | |
| Netflix Price | |
| Empty Neighborhoods | |
| 2nd Highest Salary | |
| Rolling Bank Transactions | |
| Comments Histogram | |
| Closest SAT Scores | |
| Employee Salaries | |
| Subscription Overlap | |
| Top Three Salaries | |
| Monthly Customer Report | |
| Cumulative Distribution | |
| Merge Sorted Lists | |
| Experiment Validity | |
| String Shift | |
| Top 5 Turnover Risk | |
| Like Tracker | |
| Last Transaction | |
| Find the First Non-Repeating Character in a String | |
| Bagging vs Boosting | |
| Total Spent on Products | |
| Alphabet Sum | |
| P-value to a Layman |
Synthesized from candidate reports. Individual experiences may vary.
Candidates report an initial recruiter call that mostly covers background and relevant experience. Some describe slow recruiter responsiveness and unclear communication about next steps, so it may help to ask early what each later round will cover and how it will be scheduled.
Candidates describe a hiring-manager call about their background, sometimes with ML basics or technical questions. One candidate reports the manager raised recommendation-system evaluation and target population selection, and said their background did not fit. Be ready to connect your experience to recommenders directly.
Several candidates report a technical screen with a LeetCode-style problem, such as dynamic programming or number of islands, sometimes paired with SQL or ML basics. One describes the coding as roughly LeetCode medium. Practice standard algorithm problems alongside core ML concepts.
Candidates report a virtual onsite with multiple technical interviews. Topics mentioned include ML fundamentals, MLOps system design, statistics, a binary-classification case and recommendation-system design tied to a marketing campaign. Another candidate reports two rounds on recommendation systems and broader ML with system design.
Some candidates describe a functional onsite covering collaboration, communication and leadership. Two report technical content appearing in it: a director-led recommendation case in one, and hiring-manager case studies on ranking and recommendation models with detailed follow-ups in the other. Prepare to defend tradeoffs and edge cases.
