Quick Commerce Platform – System Design Interview Preparation

Practice designing a quick commerce platform like Blinkit/Zepto. Cover dark stores, inventory management, and 10-minute delivery.

This is a full mock interview round, not a written solution to read through. The AI interviewer poses the Quick Commerce design problem, pushes back on your choices, and asks the follow-up questions a real interviewer would: how the system behaves as traffic grows, where the bottlenecks move, which consistency guarantees you are trading away, and what happens when a component fails. You get structured feedback and a score on requirement gathering, high level architecture, data modelling, scalability and communication at the end of every attempt.

Designing Quick Commerce is a recurring question in High Level Design (HLD) rounds at Google, Amazon, Meta, Microsoft, Netflix and Uber, and at high-growth product companies hiring SDE-2 and above. Working through it out loud - repeatedly, under time pressure - is what separates candidates who know the components from candidates who can defend a design.

What this question is really testing

A ten-minute delivery promise, which makes inventory accuracy per dark store the central problem rather than an afterthought.

Scale to establish before drawing anything

Hundreds of dark stores, each holding 2,000-5,000 SKUs for a 2-3 km radius.
Order-to-doorstep in 10-15 minutes, so every step has a budget measured in tens of seconds.
Inventory changes constantly and must be correct at the moment of add-to-cart, not at checkout.

Core components

Store locator – Maps a delivery address to the serving dark store. Every other decision depends on this.
Inventory service – Per-store, per-SKU counts with reservation on add-to-cart. The highest-contention service in the system.
Order service – Short-lived state machine: placed, picked, packed, dispatched, delivered.
Picker assignment – In-store task queue, optimised for pick path rather than distance.
Rider dispatch – Short radius, high frequency - a different problem from long-haul delivery.

Trade-offs to defend

Reserve on add-to-cart versus on checkout – Reserving early prevents the worst customer experience and holds stock for abandoned carts. The timeout on that hold is the number to defend.
Oversell tolerance – Strict consistency on inventory costs latency on the hottest path. Some systems oversell and substitute, which is a product decision.
Store assignment on the boundary – An address between two stores can be served by either; picking by stock rather than distance changes the promise.

Follow-up questions to expect

The last unit of an SKU is added to two carts in the same second. Who gets it?
A picker finds the shelf empty though the system says there is stock. What does the customer experience?
Demand triples in a thunderstorm. Which constraint binds first - stock, pickers, or riders?