Capital One Interview Questions — Page 22

Capital One·Software Engineer·Technical Phone Screen

Jun 2026

Capital One SWE interview with a mix of string manipulation, grid pathfinding, and a sliding window problem. The questions ranged from straightforward to genuinely tricky, and the grid one in particular had a lot of moving pieces to keep track of.

Given two equal-length strings s1 and s2, build a new string by iterating through each index i: always append s1[i], then if s1[i] equals s2[n-1-i] append that character again, otherwise append s2[n-1-i]. Return the resulting string.
A robot moves on a 2D grid with walls, healing cells, and monster cells. Starting with a fixed health value, find the shortest path from start to goal where health never reaches zero or below. Return the shortest distance or the path.
Solve a problem on an array or string using a sliding window approach to satisfy a given constraint.

“Took me a minute to parse the comparison logic since you're comparing against the mirrored index in s2, not the same index.”

View Post

Capital One·Machine Learning Engineer·Technical Phone Screen

Jun 2026

Interviewed for an ML Engineer role at Capital One and got a tree traversal problem. Pretty standard coding round, nothing too wild, but it's the kind of question where you can fumble the path-building if you're not careful.

Given the root of a binary tree, return all root-to-leaf paths as strings with node values joined by '->'.

“The traversal itself wasn't the hard part.”

View Post

Capital One·Software Engineer·Technical Phone Screen

Jun 2026

Capital One SWE interview with a coding problem around digit counting. Pretty straightforward premise but the edge cases are where it gets you.

Given an array of non-negative integers, count how many of them contain an odd number of the digit '0' in their decimal representation. The number 0 itself counts as having one zero digit.

“Spent the first minute overcomplicating it in my head.”

View Post

Capital One·Software Engineer·Technical Phone Screen

Jun 2026

Capital One SWE interview with a straightforward coding problem around bank account operations. Nothing too wild, but the implementation details trip you up if you're not careful about edge cases.

Design and implement a Bank class that supports transfer, deposit, and withdraw operations on n accounts with initial balances. Each method should validate account numbers and balances before proceeding, and return a boolean indicating success or failure. Account numbers are 1-indexed.

“Seemed simple at first and I basically started coding before fully thinking through the indexing.”

View Post

Capital One·Data Scientist·Online Assessment (OA)

Jun 2026

Capital One OA for a Data Scientist role, one SQL problem on transaction data. Nothing crazy, but the streak logic tripped me up more than I expected.

Given a transactions table with user IDs, amounts, dates, merchants, and statuses, find the longest streak of consecutive days with successful transactions for each user. Return user_id and longest_streak, ordered by streak length descending. Exclude failed or reversed transactions.

“The filter part was obvious but the consecutive-days logic took me a minute.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026

Capital One Data Scientist interview with a pricing/economics problem that felt more like a micro-econ exam than anything I expected from a tech company. The question had a few parts and required actually thinking through market behavior, not just plugging numbers.

Customers' willingness to pay increases linearly with the number of available shows, from $0 at zero shows to $20 at 40 shows, and stays capped at $20 beyond that. Sketch the WTP curve and write it as a piecewise function.
What is the marginal increase in willingness to pay for each additional show added below the 40-show threshold?
Why is it wrong to assume WTP keeps rising indefinitely past 40 shows, and what does the cap represent in terms of actual consumer behavior?

“The sketch part threw me a little because I wasn't expecting to literally draw anything on a phone screen.”

View Post

Capital One·Data Scientist·Onsite - Behavioral / Leadership

Jun 2026

Capital One onsite for a Data Scientist role, behavioral panel format with what felt like a prospective manager in the room. Three questions covering teamwork, technical depth, and an ethics curveball at the end. The ethics one is what I keep thinking about.

Tell me about the best cross-functional data project you've been part of. What was your role, how did you work with Product, Engineering, Risk, and other stakeholders, how did you navigate disagreements, and what did it actually accomplish?
Walk me through a time you used an advanced analytical or technical method to crack a hard problem. Why that method specifically, how did you validate it, and what was the business result?
Our product team is considering adding facial recognition as a login option. What ethical concerns and risks do you see, and how would you actually address them before shipping something like that?

“I had a story ready but I underestimated how much they'd probe the conflict piece.”

View Post

Capital One·Software Engineer·Technical Phone Screen

Jun 2026

Capital One applied researcher interview with a grid path problem that looks straightforward until you actually try to implement it. The alternating operator/digit constraint is what makes it interesting, and I definitely underestimated that part going in.

You're given an m×n grid where each cell holds either an arithmetic operator (+, -) or a digit (0-9). Moving only right or down from the top-left to the bottom-right, find the path that maximizes the value of the resulting arithmetic expression, where valid paths must strictly alternate between digits and operators (no two consecutive digits, no two consecutive operators).

“My first instinct was just standard DP for max path sum and I started coding that before fully registering the alternating constraint.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026

Capital One data scientist interview with a meaty experimentation case about redesigning a theme park's queue system. One question, but it had a lot of moving parts and went pretty deep.

A theme park is testing a virtual queue system aimed at cutting average wait times by 20% without tanking per-capita revenue by more than 2%. Walk through how you'd design the experiment: what metrics you'd track, how you'd handle randomization and interference between guests, how you'd size and monitor the test, and what you'd do if results come back mixed (say, waits drop more than expected but revenue falls more than the guardrail allows).

“This one sprawled in a way I wasn't ready for.”

View Post

Capital One·Machine Learning Engineer·Technical Phone Screen

Jun 2026

Capital One MLE interview focused entirely on taking a fraud detection model from notebook to production. Three distinct sections: deployment mechanics, monitoring strategy, and hitting a strict p99 latency target. Pretty technical throughout, no behavioral fluff.

How would you deploy a machine learning model to production?
What artifacts do you version when deploying an ML model, and how do you support safe rollouts and rollbacks?
After your model is live, how do you monitor it?
What specific metrics do you track across service health, data quality and drift, model performance, and business impact for a deployed fraud model?
The fraud model is served via an online API with a hard requirement of p99 latency under 50ms. How do you figure out where the time is going, and what changes would you make across features, model architecture, infrastructure, and serving to meet that target?

“I started with containerization and a REST API wrapper, which felt right, but then I rambled into CI/CD pipelines and lost the thread a bit.”

View Post