How to approach a system design interview ?

Let’s walk into the system design garden..

Q1. Clarify the requirements : What are you building ?

• The first step is to fully understand what's being asked.
• Don’t rush into designing before you know the scope. Start by asking clarifying questions to understand the problem fully.
• Consider — Who will use the system ?
• What will they use it for?
• How many users are expected ?
• What will the system do ?
• What kind of data will it handle , and how much?
• How many requests per second should it support?
• What’s the expected ratio of reads to writes?
• By asking these questions, you define the problem clearly and set yourself up to design the right solution.

Q2. Design the big picture : High-Level Architecture

• Once you understand the problem , the next step is to outline the high-level architecture.
• This means sketching out the key components and how they interact with each other.

Here’s what you need to do:

• Identify major components : Break down the system into key parts . In social media app, for e.g.. You’d have user profiles, posts, comments, and notification.
• Show connections: Explain how these components communicate — whether through APIs, databases, load balancers or caching layers.
• Justify your choices : Don't just present a design : Explain why.

Focus on key components

• After outlining the big picture , the next step is to focus on the most critical parts.
• For e.g.. If you’re building a URL shortening service, you’ll need to think about :
• How will you generate and store shortened URLs?
• What hashing technique will you use ( e.g. MD5, Base62)?
• How will you handle hash collisions?
• What kind of database will you use (SQL or NoSQL)?
• How will you store and retrieve the data efficiently ?

Plan for Scaling : Handle Growth and Traffic

• Once the core components are in place , think about how the system will scale.
• What happens when the user base grows from 1,000 to 1 million?
• Consider scalability solutions like:

Load balancer

Horizontal scaling

Caching

Database sharing

When discussing scalability, always consider trade-offs.

For eg. Caching can improve performance but might introduce complexity in keeping data up-to-date.

Quick calculation : Be ready to make quick estimates for data storage , traffic and basic numbers like latency and data sizes.

#systemdesign