← Back to Directory

Capital One interview notes, page 31 of 39

Large Enterprises

Most of these Capital One write-ups come from Data Scientists (196), then Software Engineers (93) and Product Managers (54). The most reported rounds are the technical phone screen (213), the product sense round (58), the behavioral / leadership round (41) and the system design round (22).

382 interview notes · updated Oct 2026

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One data scientist interview with a meaty pricing and content optimization case. Single question but it had a lot of moving parts and I don't think I covered all of them cleanly.
  • For an online content startup, how would you decide how many shows to produce and what subscription price to set? Walk through your optimization objective and constraints, the minimum data you'd need, how you'd estimate that data, and how your answer changes if variable cost per subscriber rises from $4 to $6.

“This question is basically four questions stapled together and I did not pace myself well.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One data science interview with a single brutally detailed question that basically asked you to run a full analysis project out loud. No warmup, no softballs.
  • Walk me through a time you used data to tackle a vague, poorly defined business problem from start to finish. They want the specific hypothesis you formed, which data sources you used and what biases you knew about, how you chose your model or statistical method, how you validated your assumptions, and a defensible way to quantify impact (not just before-and-after numbers). Also: how did you deal with data quality problems, how did you communicate uncertainty to non-technical stakeholders, and what would you do differently with 10% more data?

“This question is basically five questions wrapped in a trench coat.”

View Post

Capital One·Data Scientist·Onsite - Cross-functional / Panel

Jun 2026
Capital One data scientist interview with a pretty gnarly cross-functional question that felt more like a program manager stress test than anything data-related.
  • Walk me through a project where you had to coordinate across at least three departments with conflicting goals. What was the business objective, where exactly did people disagree, and what framework did you use to break deadlocks? Include the artifacts you made, how you handled scope and quality trade-offs, and what the measurable outcome was. Then tell me what you'd do differently if one constraint changed.

“This question is long and they mean all of it.”

View Post

Capital One·Software Engineer·Onsite - System Design / Architecture

Jun 2026
Capital One system design round focused entirely on building a rock-solid account balance system. Dense topic, lots of ground to cover, and I kept feeling like I was one follow-up away from losing the thread.
  • Design a highly reliable account balance system that supports deposits, withdrawals, and transfers, with strong guarantees that no money is ever lost or duplicated.

“This is basically a distributed systems gauntlet dressed up as a single question.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One data scientist interview with a meaty case question that blended experiment design, business judgment, and break-even math all in one prompt. Felt more like a consulting case than a typical DS screen.
  • A restaurant is considering joining a coupon program. What concrete factors and metrics would you evaluate before signing on, how would you design an experiment to estimate incremental profit impact, and what is the minimum share of coupon tables that must be truly incremental for the program to break even, given specific baseline and observed numbers?

“This was a three-part monster and I didn't pace myself well.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One data scientist interview, looked like a technical screen focused entirely on ML fundamentals. Four questions, all written, all requiring actual depth. Not a vibe check at all.
  • When would you choose MAE over MSE as a regression loss? Walk through the robustness to outliers, how gradients behave near zero, and give a concrete scenario where MSE fails but MAE holds up.
  • You have a binary classifier with 1% positive prevalence, ROC-AUC of 0.90, and PR-AUC of 0.25. Which metric matters more here and why? How can ROC look strong while PR stays weak?
  • For a neural network binary classifier, what output activation and loss would you use? How do class weights and focal loss each change how the model treats positive versus negative examples during training?
  • If the business needs precision at or above 0.50, how do you pick a probability threshold using validation data without introducing optimistic bias?

“This one I actually felt okay about.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One Data Scientist interview that leaned harder into financial modeling than I expected. The core question was a renewable energy plant ROI case, and you basically had to work through the math live to figure out whether the investment made sense.
  • A renewable energy plant has an average output of 1,000 kWh, a max annual capacity of 8.8 million kWh, a monthly land lease of $5M, fixed annual costs of $25M, variable cost of $20/kWh, and a selling price of $40/kWh. The initial investment is $400M. What is the minimum annual kWh output the plant needs to generate and sell to hit a 10% annual ROI? And given that number relative to the 8.8M kWh capacity ceiling, would you approve the investment?

“I knew the ROI formula but blanked for a second on how to wire up all the cost components cleanly.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One data scientist interview with a single sprawling case study about a grocery loyalty card program. The whole thing was basically one giant causal inference and experimentation design question broken into five parts. Dense, but kind of interesting if you like that stuff.
  • A national grocery chain launches a free loyalty card program on January 1. Using 18 months of household-level transaction data, enrollment dates, coupon redemptions, and acquisition/servicing costs, define the causal estimand and write out the exact incremental profit formula including gross margin, discount cannibalization, coupon funding, acquisition cost, servicing cost, and fraud/breakage.
  • Propose an identification strategy for estimating the loyalty program's causal effect. Write out a difference-in-differences specification including the outcome variable, treatment indicator, and fixed effects, and list the assumptions you would test such as parallel trends, composition stability, seasonality, and event timing.
  • What guardrail metrics would you set for this loyalty program evaluation, and how would you detect and mitigate selection bias given that enrollment is voluntary?
  • Walk through how you would determine the required sample size and statistical power for this experiment. What minimum detectable effect would you target, and what inputs do you need to derive it?
  • Design an A/A test and a pre-launch placebo difference-in-differences to validate your evaluation framework before the program goes live. Be specific about time windows, cohort definitions, and the decision rule you would use to ship or roll back.

“I spent way too long on the profit formula and not enough time being precise about the estimand.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One data scientist case interview, heavy on product strategy and analytics for a credit card partnership evaluation. The whole thing was basically one giant open-ended case about whether to sign two co-branded partnerships, and they expected you to go pretty deep on the quantitative side.
  • How would you define 'high value' cardholder quantitatively for the purpose of evaluating a co-branded credit card partnership?
  • What are the top factors you would analyze before signing a partnership with a home-sharing platform and a big-box retailer, and how would you justify prioritizing them?
  • Design a measurement plan for evaluating whether a co-branded card partnership is successful, including your primary KPI, guardrails, and experiment design.
  • How would you detect and protect against credit card gamers or churners in a promotional offer tied to this partnership?
  • What data would you need and how would you segment your analysis across new vs existing cardholders and peak vs off-peak periods?
  • What go/no-go thresholds would you set for deciding whether to proceed with the partnership, and how would you frame payback period vs IRR?

“I started with CLV and the interviewer pushed back immediately asking for specifics.”

View Post

Capital One·Data Scientist·Technical Phone Screen

Jun 2026
Capital One data scientist interview with a multi-part burger profitability case. The math gets layered fast once you introduce a second product line with its own cost structure, and they want both the formula and the number.
  • In Year 2, a Vegan burger option is added with fixed costs of $60m/year in training and $2.25m/month in supplier costs. Prices and variable costs stay the same (Regular: $4 price, $1 cost; Vegan: $4 price, $2 cost), and the sales mix is Vegan to Regular at 2:3. What total number of burgers do you need to sell in Year 2 to maintain the same profit margin as Year 1? Give both the general formula and the final number.

“This one is where I started second-guessing myself.”

View Post