Autocoder

Intuit Interview Questions and Answers for SDE-1 Roles

Last updated 2026-11-19

Intuit builds financial software, so good answers show care with money, data accuracy and customers, besides the usual data structures and SQL. This page works through technical and behavioural questions with that lens.

The questions are practice in the style of topics reported by candidates, not a record of past interviews.

Entry roleSoftware Engineer 1 (SDE-1)
Reported testOnline coding: DSA, SQL, scripting
Reported interviewsTwo technical, one behavioural or manager
Official sitejobs.intuit.com

How the rounds are commonly reported

Intuit does not publish a full fresher process in the sources used here. Candidate reports, which are not official, describe an online coding assessment (reports vary: three to five data structures, SQL or scripting questions, and a 2021 SDE-1 account mentioned HackerEarth with four problems in about 90 minutes), then typically two technical interviews of around 45 minutes each, then a behavioural or hiring-manager conversation. Topics reported include data structures and algorithms, SQL and basic scripting. The usual entry title is Software Engineer 1 (SDE-1).

Intuit's own blog advice on acing its software engineer interview is a useful companion to these reports. Since the reports differ, confirm the stages for your own application with your recruiter.

Related practice

For coding practice in the style of the assessment, use Intuit coding round questions. For the process see how to crack the Intuit interview.

Hiring status

Intuit has no sourced hiring statements in the records used for this site, and there is no Intuit page under hiring news here. For current roles and dates, use Intuit's own careers site, jobs.intuit.com.

Intuit Interview Questions and Answers for SDE-1 Roles: questions with worked answers

  1. 1

    Why should money never be stored as a floating-point number, and what do you use instead?

    easy

    Binary floating point cannot represent most decimal fractions exactly, so 0.1 + 0.2 is not exactly 0.3, and errors accumulate across many transactions. Store money as an integer number of the smallest unit (paise or cents) or use a decimal type such as BigDecimal in Java or Decimal in Python, and set an explicit rounding mode, for example half-even for tax calculations. Also store the currency alongside the amount.

    python
    from decimal import Decimal, ROUND_HALF_EVEN
    print(0.1 + 0.2)                       # 0.30000000000000004
    print(Decimal('0.1') + Decimal('0.2')) # 0.3
    tax = (Decimal('199.99') * Decimal('0.18')).quantize(Decimal('0.01'), rounding=ROUND_HALF_EVEN)
    print(tax)                             # 36.00
    java
    import java.math.*;
    class Money {
        public static void main(String[] args) {
            BigDecimal a = new BigDecimal("199.99");
            BigDecimal tax = a.multiply(new BigDecimal("0.18")).setScale(2, RoundingMode.HALF_EVEN);
            System.out.println(tax); // 36.00
        }
    }
  2. 2

    Design an expense tracker that automatically categorises transactions.

    medium

    Clarify inputs (bank feeds, manual entry) and goals (monthly summaries, budgets). Ingest transactions, normalise merchant names, then categorise with rules first (merchant to category map, keywords) and a learned model for the rest, with user corrections stored as overrides that win next time. Store transactions in a relational table keyed by user and date with a categories table, precompute monthly aggregates, and treat the data as sensitive: encrypt it, and limit access. Mention de-duplication of the same transaction arriving from two sources.

  3. 3

    Write a SQL query for the top three spenders in each category.

    medium

    Aggregate spend per user and category, then rank within each category using a window function: SELECT category, user_id, total FROM (SELECT category, user_id, SUM(amount) AS total, DENSE_RANK() OVER (PARTITION BY category ORDER BY SUM(amount) DESC) AS r FROM expenses GROUP BY category, user_id) t WHERE r <= 3. Explain why DENSE_RANK keeps ties, whereas ROW_NUMBER breaks them arbitrarily, and that you cannot filter on the window result in the same SELECT, hence the subquery.

  4. 4

    How do you make a payment or filing-submission API safe to retry?

    medium

    Make it idempotent: the client sends a unique idempotency key with the request, the server stores the key and the response in the same transaction as the work, and a repeat with the same key returns the stored result without redoing it. Use a unique constraint on the key to handle two simultaneous retries. Distinguish transient errors (retry with backoff) from validation errors (do not retry). Without this, a timeout after success causes a double charge.

  5. 5

    Explain ACID with a double-entry bookkeeping example.

    medium

    In double-entry accounting every transaction debits one account and credits another by the same amount. Atomicity means both postings happen or neither does, consistency means the total of debits equals credits after every commit, isolation means two concurrent transfers do not both spend the same balance, and durability means a committed posting survives a crash. Use a transaction with a row lock or a serialisable isolation level for the balance check and update, and never update the two sides in separate transactions.

  6. 6

    How would you design pagination for an API that lists transactions?

    medium

    Offset pagination (LIMIT and OFFSET) is simple but slow for deep pages and unstable if rows are inserted while paging. Cursor pagination uses the last seen sort key, for example (date, id), and queries for rows after it, which is fast with an index and stable. Return a cursor token, a page size cap and a flag for more results. Add a composite index matching the sort order so the database does not sort in memory.

  7. 7

    When would you pick an array over a hash map for counting?

    easy

    When the keys are small integers or a fixed small alphabet, such as lowercase letters or values from 0 to n, an array is faster and uses less memory because it avoids hashing and has better cache locality, and lets you iterate in key order for free. A hash map is better when keys are sparse, unbounded or non-numeric. State the memory cost: an array wastes space if the key range is huge and mostly empty.

  8. 8

    How do you detect duplicate transactions in a stream?

    hard

    Define a key that identifies a duplicate, for example (account, amount, merchant, timestamp bucket) or a provider's transaction id. Keep recently seen keys in a hash set, or a Bloom filter for a memory-light approximate check followed by an exact check on positives, with entries expiring after the window in which duplicates can occur. Discuss false positives, which would block a legitimate repeated purchase, and decide the policy for suspected duplicates (flag for review, not delete).

  9. 9

    Tell me about a time you put a customer's interest first.

    medium

    Intuit's interview advice stresses understanding the customer. Pick a story where you changed a decision because of a user's need: you noticed a confusing step, talked to users or support, and simplified it, or you chose correctness over speed because wrong results would hurt people. Quantify the result, such as fewer support questions or faster completion. Use the structure in the behavioural interview guide.

  10. 10

    Tell me about a time you found a mistake in your own work after it shipped.

    medium

    Describe what went wrong, how you found out, how quickly you told the people affected, how you fixed it and what you changed to prevent a repeat, such as a test or a review checklist. For software that handles people's money or tax data, interviewers value honesty and a clear process more than a story with no mistakes.

  11. 11

    Describe a project in which you had to work with ambiguous requirements.

    easy

    Explain how you clarified the goal (asked who the user was and what success meant), made your assumptions explicit, built a small first version to get feedback and adjusted. Give the outcome and what you learned. Be clear about decisions you made yourself, not those made by the team in general.

  12. 12

    Walk me through your resume project and the trade-offs you made.

    medium

    Be ready to explain the goal, architecture, data model and why you chose them. Pick two trade-offs, such as a relational database versus a document store or synchronous versus queued processing, and explain what you gained and gave up. Expect follow-ups about scale and failure, such as what breaks at ten times the users. Answer honestly where you did not test something.

Frequently asked questions

What is the Intuit SDE-1 process?

Candidate reports describe an online coding assessment, two technical interviews and a behavioural or manager conversation. These are not official.

Does Intuit ask SQL questions?

Reports mention SQL in the online assessment, so prepare joins, grouping and window functions.

Which platform does the Intuit test use?

A 2021 candidate report cited HackerEarth. Platforms change, so check your invitation.

Where can I find Intuit's openings?

On Intuit's own careers site, jobs.intuit.com.

Sources

Related guides

More in Company Interview Questions

See all in this section