Zepto Interview Questions and Answers
Last updated 2026-11-19
Zepto is a quick-commerce company built around dark stores and very short delivery promises, so its engineering questions, as candidates describe them, mix fast data structures problems with design questions about inventory at store level, order picking, rider assignment and schema design. This page has practice questions with worked answers across those, plus HR.
They are practice questions on the topics the rounds are commonly reported to cover, not a record of any real interview. For code see Zepto coding round questions; for the process see how to crack the Zepto interview.
| Typical rounds | DSA, LLD plus HLD, hiring manager (commonly reported) |
|---|---|
| Roles | SDE-1 |
Rounds as commonly reported
Recent candidate reports describe a DSA or problem-solving round where optimised solutions are expected quickly, then a low-level and high-level design round that includes a database schema and APIs, and then a hiring-manager round on fit and experience.
Zepto does not publish its rounds, so treat this as informal and expect variation.
Hiring status
A HRKatha report says Zepto's permanent headcount fell 3.6 percent to 5,212 in FY26, its technology headcount fell 15.6 percent, and permanent-employee attrition was 51.28 percent. See Zepto hiring for more, and apply through the careers page at zeptonow.com/careers.
How to prepare
Draw a dark store: shelves, pickers, a packing counter and a rider handoff. Then list every state a single order passes through and the data each step needs. Doing this once makes both design questions much easier.
Zepto Interview Questions and Answers: questions with worked answers
1 How would you track inventory per dark store accurately?
mediumKeep stock per SKU per store as the source of truth in a database, updated by events: inbound receipts, picks, cancellations, damage and cycle counts. Reserve stock when an order is placed so two customers do not claim the last item, and release on cancellation. Reconcile the system count with physical counts regularly, and show the customer only items with reserved-safe availability in their store.
2 Design the order picking assignment inside a dark store.
hardOrders arrive on a queue. A picker service batches orders and assigns each to a picker by current load, using a route over shelf locations so items are picked in a sensible order. Track item-level states (pending, picked, missing), handle substitutions by offering the customer a swap, and measure pick time. Priority goes to orders closest to their promised time.
3 How is the delivery time promise calculated?
hardAdd expected pick and pack time (from store load), the wait for a rider, and travel time from the store to the address. Cap serviceability to a radius, show the estimate on the product page from a cached store-level value, and recompute at checkout. When load rises, stretch the estimate or pause ordering for the store rather than breaking the promise.
4 Design a database schema for products, stores and stock.
mediumTables: Product (global), Store, StoreProduct (store, product, price, status), Stock (store, product, quantity, reserved), Order, OrderItem and an inventory event log. Index on store and product for stock lookups. Keep quantity changes in conditional updates, and separate catalogue data (changes rarely, cached widely) from stock (changes constantly).
5 Design low-level classes for a rider assignment module.
mediumClasses: Rider (id, location, status), Order, Assignment, and a Dispatcher using an AssignmentStrategy interface (nearest, least-loaded). The dispatcher takes a lock or uses compare-and-set on the rider's status, so one rider is not given two orders unless batching allows it. Add events for assignment accepted or timed out, and a driver class to demonstrate a few orders.
6 How would you design search with local store availability?
hardMaintain a search index of the global catalogue and, at query time, filter or boost by items available in the user's store, fetched from a fast store-level availability cache. Rank by relevance and likelihood of purchase. Handle out-of-stock items by showing them last or suggesting substitutes, and keep the availability cache fresh from stock events.
7 What is a distributed lock and when is it a bad idea?
mediumIt lets multiple processes agree that only one holds a resource, commonly implemented with Redis or a coordination service with an expiry. It is risky because expiry can allow two holders after a pause, so use fencing tokens, and avoid it when a database conditional update achieves the same thing. Prefer the simplest correct mechanism.
8 How do you design the API for placing an order?
mediumPOST /orders with an idempotency key, the cart reference, address and payment choice. The server validates serviceability, price and stock, creates the order in a pending state, reserves stock, and returns an order id, with payment confirmation arriving separately. Use clear error codes for sold out and unserviceable and keep the response small.
9 Explain database transactions and when you need them in an order flow.
mediumA transaction makes a group of changes atomic. Use one to create the order row, its items and the stock reservation together, so you never have a reservation without an order. Keep transactions short, and for steps across services such as payment, use a saga with compensating actions rather than a distributed transaction.
10 How do you handle a store that suddenly receives ten times normal orders?
mediumThrottle: cap orders accepted per store per minute, extend the delivery estimate, and temporarily hide slow-picking categories. Autoscale stateless services, and shed non-essential work. After the event review whether the demand forecast and staffing tool failed and fix that.
11 Tell me about a time you worked under extreme time pressure.
easyUse STAR. Give the deadline and the stakes, how you picked what to do first, what you cut or postponed on purpose, and how you communicated. Close with the result and the follow-up you did afterwards to restore quality.
12 Why Zepto, given the pace and the changing headcount?
easyGive a real reason such as the engineering interest of fast logistics and store-level systems. Be honest that a high-intensity environment is part of the draw and that you have thought about it. Reference a project where you delivered under constraints.
Frequently asked questions
Does Zepto ask both LLD and HLD?
Recent candidate reports describe a design round covering low-level and high-level design, including schema and APIs.
How fast do I need to solve DSA problems?
Reports say optimised solutions are expected quickly, so practise stating the best approach early.
Is the process the same for every team?
Zepto does not publish it, so it varies. Confirm with your recruiter.
Where can I read sourced Zepto hiring news?
See Zepto hiring and the company's careers page.
Sources
Related guides
More in Company Interview Questions
- Zerodha Coding Round Questions With Solutions
- Zerodha Interview Questions and Answers
- Zoho Coding Round Questions with Solutions
- Zoho Interview Questions and Answers
- Zomato Coding Round Questions with Solutions
- Zomato Interview Questions and Answers
- ZS Associates Coding and Case Logic Questions
- ZS Associates Interview Questions with Answers
- Accenture Coding Round Questions with Solutions
- Accenture Interview Questions for Freshers
- Adobe Coding Round Questions with Solutions in Python and Java
- Adobe Interview Questions and Answers for SDE Roles