Capital One Interview Questions — Page 20

Capital One·Data Scientist·Onsite - Behavioral / Leadership

May 2026

Capital One data scientist behavioral round, basically three big stories with a lot of specifics required. The bar for detail was higher than I expected, not just 'tell me about a win' but dates, numbers, what your manager said, and what you'd change.

Walk me through your most consequential accomplishment from the past two years. Include the context, goal, constraints you faced, what you specifically did, measurable before/after outcomes, and how your manager or others validated it.
Describe a non-trivial failure from the last two years. What early warning signs did you miss, what was the actual root cause, what did you change after, and how did your manager respond?
Tell me about a time you helped others succeed. Who were the stakeholders, what conflicts or trade-offs came up, what did you delegate or teach, and what was the measurable impact on the team or business?

“I had a solid story but fumbled the metrics part.”

View Post

Capital One·Product Manager·Onsite - Product Sense / Strategy

May 2026

Capital One PM interview, got a product design question about a fitness app. Pretty open-ended, which I wasn't totally prepared for coming from a more structured background.

Design a fitness app from scratch.

“I jumped straight into features without anchoring on a user segment first, which I regretted about two minutes in.”

View Post

Capital One·Machine Learning Engineer·Technical Phone Screen

May 2026

Capital One ML engineer interview, one coding question that looked deceptively simple on the surface. The algorithm was described in plain English rather than given as a standard problem, which threw me off more than I expected.

Implement an algorithm from a verbal description: given a non-negative integer array, initialize a result variable to 0. Find the first non-negative number x from the left, then compare subsequent elements to x. If a subsequent element is smaller, add x to the result and restart from step 1. Otherwise, subtract x from that element. Return the result when the array is exhausted.

“The description sounds clear when you read it back but in the moment I kept second-guessing what 'subtract x' meant for the array state and whether I was mutating in place or working on a copy.”

View Post

Capital One·Data Scientist·Technical Phone Screen

May 2026

Capital One data scientist interview with a business strategy angle I wasn't expecting. They put a hypothetical partnership scenario in front of me and wanted qualitative thinking, not numbers.

If a bank were considering partnering with a ride-sharing company, what qualitative factors would you evaluate before committing to the partnership?

“I went straight to data and metrics instinctively, which was probably the wrong move for a qualitative question.”

View Post

Capital One·Data Scientist·Technical Phone Screen

May 2026

Capital One Data Scientist interview with a real estate profitability case. The whole thing was basically a financial modeling exercise dressed up as a business problem, which I wasn't expecting going into a DS role interview.

For a 100-unit apartment project with 20% vacancy, a mortgage near 4.6%, $600k in annual interest, $400k in fixed costs, and variable costs at 10% of rent, what financial and operational factors need to be weighed when deciding on a rent price?
Given 80% occupancy on the 100-unit property, what is the minimum monthly rent per unit needed to break even or hit a target profit?
If rent per unit is dropped to $1,200 per month, what occupancy rate would be required to maintain the same annual profit as before?

“I started rattling off the obvious stuff like mortgage and fixed costs, but then blanked on how to frame variable costs relative to rent.”

View Post

Capital One·Data Scientist·Technical Phone Screen

May 2026

Capital One data scientist interview, SQL round. One question, but it had a lot of moving parts and the constraints made it trickier than it looked on the surface.

Given a schema with customers, orders, order_items, and products tables, write a single SQL statement that returns the product category with the highest total delivered revenue in August 2025 for each region. Revenue is qty times unit_price, filtered to delivered orders only. Output region, category, and total_revenue. Break ties by picking the lexicographically smallest category. You must use exactly one subquery or one CTE, and no window functions.

“The no-window-functions constraint is what got me.”

View Post

Capital One·Machine Learning Engineer·Technical Phone Screen

May 2026

Coding round for an ML Engineer role at Capital One. Two algorithm problems, one straightforward and one that felt like it came out of nowhere.

Rotate an n×n matrix 90 degrees clockwise. Extra space is allowed.
Given a 1-D grid with lamps placed at certain positions, find the point or points that are illuminated by the maximum number of lamps.

“Fairly standard matrix problem.”

View Post

Capital One·Product Manager·Onsite - Product Sense / Strategy

May 2026

Capital One PM interview with a product strategy question about their shopping extension. Pretty open-ended, which I wasn't fully prepared for.

How would you expand Capital One Shopping as a product?

“I went straight into feature ideas without anchoring on who the current users actually are or what problem the expansion would solve.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026

Capital One Data Scientist interview with a business case question about an amusement park launch. Pretty much a full profitability exercise dressed up as a data role question, which I wasn't totally expecting.

An amusement park operator is considering launching a new park. Estimate annual revenues across ticket sales, food and beverage, merchandise, and parking. Then estimate fixed and variable costs including construction, operations, maintenance, and staffing. Based on that, calculate expected profit and give a recommendation on whether to proceed.

“I spent way too long on the revenue side and then rushed the cost breakdown.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026

Capital One data scientist loop, heavy on the technical side. Four questions back to back covering SQL reconciliation logic, ETL architecture, visualization justification, and a Python vs R debate. Felt more like a take-home that someone decided to do live.

Write a single SQL query that produces a payment-level reconciliation report across two financial ledgers, handling FX conversion using the most recent rate on or before each payment date, matching transactions by user, amount tolerance of $0.01, and time window of 48 hours, then classifying each payment as matched, late (within or beyond 48h), amount mismatch, or missing in one ledger. Use window functions to break ties.
Outline an idempotent daily ETL pipeline in Python that partitions by event date, maintains a 72-hour rolling backfill window for late arrivals, writes a reconciliation snapshot with a deterministic primary key, and guarantees exactly-once downstream effects. How do you handle merging late_beyond_48h corrections without duplicating rows?
Generate a pie chart of match_status counts using matplotlib or plotly, then explain why a bar chart with percentages and 95% confidence intervals is a better choice for this use case. Name the specific Python packages you would use.
Defend your choice of Python over R for this reconciliation pipeline, being specific about library comparisons (pandas, pyarrow, duckdb vs data.table, arrow, dbplyr), deployment and runtime concerns, and how Python fits into a modern data stack.

“This was the one that hurt.”

View Post