Autocoder

Meesho Interview Questions and Answers

Last updated 2026-11-19

Meesho serves value-conscious shoppers across small towns, many on low-end phones and patchy networks, through resellers and suppliers. That context shapes the practice questions below: lightweight APIs, catalogue and image scale, a cart that survives flaky connections, and cost-aware design. Each question has a worked answer.

They are practice questions on the topics the rounds are commonly reported to cover, not a record of what any candidate was asked. For code problems see Meesho coding round questions; for the process see how to crack the Meesho interview.

Typical roundsOnline DSA test, machine coding, technical plus culture fit (commonly reported)
RolesSDE-1, SDE intern

Rounds as commonly reported

Candidate reports describe an online DSA test with three medium to hard problems in about 90 minutes, a machine-coding or low-level design round with tasks such as a shopping cart or cab booking, and a technical plus culture-fit round covering projects and behaviour.

Meesho does not publish its rounds, so treat the above as an informal guide that varies by role.

Hiring status

One sourced fact: Meesho's Scripted By Her 2.0 (2026) is a women-only hackathon with Rs 2.25 lakh in prizes, and finalists may get interviews for a six-month internship or full-time roles, with no guaranteed pre-placement offer stated (source: Internshala listing). See Meesho hiring for more, and apply through the careers page at meesho.io/jobs.

Mindset for the design questions

Say the constraint first: low bandwidth, low memory, intermittent connectivity, price sensitivity. Interviewers in this setting often want to see that you design for the user you are serving, not for an ideal one.

Meesho Interview Questions and Answers: questions with worked answers

  1. 1

    How would you design the product catalogue and image delivery for low-bandwidth users?

    medium

    Store images in object storage, generate several resolutions and formats such as WebP at upload, and serve through a CDN with the client requesting a size matched to its screen and network. Lazy load below the fold, use placeholders, and cache aggressively on device. Keep list-page responses small by returning only fields the card needs, and compress responses.

  2. 2

    Design a shopping cart that tolerates network failures.

    medium

    Keep the cart on the server keyed by user, with the client holding a local copy and queuing changes while offline. Each change carries an operation id so a retry is idempotent, and conflicts resolve by applying operations in order. At checkout revalidate price and stock on the server and tell the user clearly about changes.

  3. 3

    Design a cab booking system at the class level.

    medium

    Entities: Rider, Driver, Cab, Trip, Location. A Dispatcher finds the nearest available cab through a pluggable MatchingStrategy, and a Trip has states Requested, Assigned, Started, Completed, Cancelled. Pricing is a separate strategy. Marking a driver assigned must be atomic so two riders cannot share one cab. Provide a driver class that books, starts and completes a trip.

  4. 4

    How would you build search for a catalogue with millions of products?

    hard

    Use a search engine with an inverted index over title, category and attributes, with analyzers for spelling variants, transliteration and local language terms. Rank by text relevance, conversion and availability. Provide filters through facets, cache popular queries, and update the index from product change events.

  5. 5

    How do you keep prices consistent between the listing, cart and order?

    medium

    The listing can tolerate a slightly stale price, but the order must not. Treat the price service as the source of truth, store the price on the cart line when added, but re-fetch at checkout and again at payment. Snapshot the final price on the order so later changes never alter it.

  6. 6

    What is a cache stampede, and how do you prevent it?

    medium

    When a hot key expires, many requests miss together and hit the database at once. Prevent it with a lock so only one request recomputes while others wait or serve stale data, with randomised expiry (jitter) so keys do not all expire together, and by refreshing popular keys before they expire.

  7. 7

    Explain how you would design reseller commission calculation.

    hard

    Capture each sale as an immutable event, and compute commission in a batch or stream job from rules (percentage by category, caps, returns). Hold the commission as pending until the return window ends, then confirm it in a ledger. Reversals for returns are new entries, not edits, so the history is auditable.

  8. 8

    Difference between horizontal and vertical scaling, and which would you pick first?

    easy

    Vertical scaling adds resources to one machine and is simple but has a ceiling and a single point of failure. Horizontal scaling adds machines behind a load balancer and needs stateless services and shared storage. Start vertical for simplicity if load is modest, and design stateless services early so horizontal scaling is possible.

  9. 9

    What are indexes, and how would you choose them for an orders table?

    medium

    Index the columns used in filters and joins: user id with created time for order history, order id for lookup, and status with created time for operational queries. Use composite indexes in the order of equality columns then range columns. Check query plans, and avoid indexing columns that rarely filter or that change constantly.

  10. 10

    Tell me about a time you worked with limited information or resources.

    easy

    Use STAR. Describe the constraint, how you decided what to do first, what you cut or simplified, and how you checked it worked. Mention frugality as a strength: delivering value without waiting for ideal conditions.

  11. 11

    Why Meesho?

    easy

    Anchor your answer in the audience and the engineering challenges that follow: scale of small-town users, cost-sensitive design, and trust between sellers and buyers. Connect it to something you built or studied, and say what you would like to learn.

  12. 12

    How do you handle feedback you disagree with?

    easy

    Listen fully, ask for specific examples, check whether the underlying concern is valid even if the proposed fix is not, and respond with evidence. Give an example where you changed your approach, and one where you explained your reasoning and the other person agreed or you agreed to measure.

Frequently asked questions

What is the Meesho online test like?

Candidate reports describe three medium to hard DSA problems in about 90 minutes. Check your own invite for the real format.

Is machine coding part of every Meesho interview?

Reports mention it for SDE-1, with tasks such as a cart or cab booking. It varies by team.

Does Meesho hire through hackathons?

The sourced listing says finalists of Scripted By Her 2.0 may be interviewed for internships or full-time roles. See Meesho hiring.

How should I prepare for culture-fit?

Prepare two or three real stories on ownership, resourcefulness and feedback, and keep them honest and specific.

Sources

Related guides

More in Company Interview Questions

See all in this section