American Express - interview guide

American Express interview questions - a complete guide

A mix of SQL, Python, probability and guesstimates has been reported for American Express analyst roles. Here is how to prepare for every kind of question.

Language:EnglishHinglish

About American Express

American Express is a global company in payments, credit cards and financial services. In India, freshers usually join analytics and data science roles - such as credit risk, fraud and decision science teams.

This guide covers 13 American Express interview questions with answers - what each question checks and how to answer it - plus the American Express interview process, round by round, a 7-day preparation plan for freshers, tips and the mistakes to avoid.

American Express hiring process

  1. Step 1: Coding / online test

    A coding test, or an aptitude and SQL test.

  2. Step 2: Technical interviews

    Python, SQL, pandas, guesstimates, probability, puzzles - ML questions for data science roles.

  3. Step 3: Resume and HR

    Projects and fit.

American Express does not publish its fresher hiring process in detail (as of our check). This information is based on candidate reports and varies by role, city and drive - read the official job posting before you apply.

Sources (checked 16 October 2026): American Express Careers. Hiring processes change - read the official notice before you apply.

American Express interview questions and answers

SQL and Python

  1. 1.SQL: each customer's total spend per month.

    What they're checking
    Everyday analytics work.
    How to answer
    GROUP BY customer_id and month (extracted from the date), SUM(amount). Then ORDER BY.

  2. 2.How do you use groupby in pandas?

    What they're checking
    Handling data in Python.
    How to answer
    df.groupby('city')['amount'].sum() - group, pick a column, aggregate. Use .agg() for several aggregates.

  3. 3.What would you do with missing values?

    What they're checking
    Data cleaning.
    How to answer
    First understand why they are missing; then drop them, fill with the median/mode, or use a separate category - depending on the data and its use.

Probability, guesstimates and puzzles

  1. 4.Two cards are drawn from a deck (without replacement). What is the probability both are aces?

    What they're checking
    Probability - reported on campus.
    How to answer
    4/52 × 3/51 = 1/221. Solve it while stating the steps.

  2. 5.Guesstimate: how many credit cards are active in Bengaluru?

    What they're checking
    Structured thinking.
    How to answer
    Population → working adult population → % with a card → average cards per person. State every assumption.

  3. 6.Puzzle: 8 balls, one heavier - how many weighings at minimum?

    What they're checking
    Logical thinking.
    How to answer
    Two: weigh 3 against 3; if equal, weigh one of the other 2 against the other; if not, weigh 1 against 1 from the heavier group of 3.

Data science (DS roles)

  1. 7.What is overfitting and how do you prevent it?

    What they're checking
    ML basics.
    How to answer
    The model memorises the training data and does badly on new data. Prevent it with more data, a simpler model, regularization and cross-validation.

  2. 8.In fraud detection, which matters more - precision or recall?

    What they're checking
    Metrics combined with business thinking.
    How to answer
    Usually recall (catching more fraud), but too many false alarms annoy customers - discuss the balance between the two and the costs.

  3. 9.Tell me about an analytics project you did - what insight did you find?

    What they're checking
    Resume deep dive.
    How to answer
    The data, the approach, the insight and its impact - with a number.

More questions

  1. 10.What is the difference between a window function and GROUP BY in SQL?

    What they're checking
    Depth in SQL.
    How to answer
    GROUP BY collapses rows into one; a window function keeps every row and adds an aggregate alongside (like a running total).

  2. 11.How would you design an A/B test?

    What they're checking
    Analytics.
    How to answer
    Split users randomly into two groups, fix one metric, think about the sample size, and check the result's statistical significance.

  3. 12.When do you use logistic regression?

    What they're checking
    ML basics (DS roles).
    How to answer
    For yes/no outcomes (fraud or not) - it gives a probability.

  4. 13.What does credit risk mean, simply?

    What they're checking
    Understanding the domain.
    How to answer
    The risk that a customer won't repay a loan - judged from income, past record and spending patterns.

7-day preparation plan for the American Express interview

  1. Day 1: SQL - joins, GROUP BY, window functions.
  2. Day 2: pandas - cleaning, groupby, merge.
  3. Day 3: Probability - 10 questions.
  4. Day 4: Guesstimates - three, out loud.
  5. Day 5: Puzzles - five classics.
  6. Day 6: ML basics and your project (DS roles).
  7. Day 7: A mock interview.

Preparation tips

  • Practise SQL daily - GROUP BY, joins, window functions.
  • Probability basics and the structure of guesstimates.
  • Data cleaning and groupby in Python pandas.
  • For a DS role, ML basics and metrics.

Mistakes to avoid

  • Not stating assumptions in a guesstimate.
  • Forgetting the difference between 'with' and 'without replacement' in probability.
  • Not linking metrics to the business.

Frequently asked questions

What is asked in the American Express analyst interview?

Commonly SQL, Python/pandas, probability, guesstimates and puzzles; ML for DS roles.

Is there a coding test at American Express?

Many reports name a coding or online test as the first step.

Which roles does American Express offer freshers?

Commonly analytics and data science roles - such as credit risk, fraud and decision science.

How long is the American Express process?

Some reports say about two weeks - it depends on the role.

Similar companies - Finance and investment banks

More company guides

Read next

On a live call, Cuelyte AI hears questions like these and has an answer ready in about a second, built from your resume and notes. Client calls: how Cuelyte AI helps

Ready for your American Express interview?

Practise these questions with Cuelyte AI: add your resume, type a question and get a clear answer in your own words - in English, Hindi or Hinglish. Free to start.

Start free - no card neededSee pricing