System Design Interview Questions
Real system design questions asked at top tech companies, covering distributed systems, scalability, caching, databases, microservices, and more.
System design is the most consistently tested competency in senior backend, cloud, and SRE interviews. Unlike algorithm questions, design questions are open-ended: there’s no single correct answer, and interviewers are evaluating your reasoning process as much as the solution.
What system design interviews actually test
Breadth of knowledge
Load balancers, caches, message queues, databases, CDNs, API gateways, service meshes.
Judgment under constraints
Trade-offs like strong consistency vs. availability, cost vs. performance — knowing why, not just that both exist.
Scalability thinking
Identifying bottlenecks, designing for 10x load, when to shard, cache, or go async.
Communication
Explaining reasoning clearly in real time, not just stating a conclusion.
System design questions on CurseAndCure
How would you design a scalable microservices architecture for an online food delivery platform?
Topic: Microservices, service decomposition, API gateway, event-driven architecture
Read answer →
How do you design a real-time notification system, considering scalability and message delivery guarantees?
Topic: WebSockets, message queues, fan-out, at-least-once delivery
Read answer →
How do you design a disaster recovery strategy for a mission-critical application on Google Cloud, considering RTO and RPO?
Topic: DR strategy, RTO, RPO, failover, multi-region
Read answer →
How do you design secure and performant hybrid cloud connectivity?
Topic: VPC peering, VPN, Direct Connect, hybrid architecture
Read answer →
How would you design a comprehensive cloud monitoring strategy for a distributed application?
Topic: Observability, metrics, logs, traces, cost management
Read answer →
Core system design topics to master
- Caching — when to cache, invalidation strategies, eviction policies, distributed caches.
- Databases — RDBMS vs. NoSQL, ACID, indexing, read replicas, sharding, OLTP vs. OLAP.
- Message queues — Kafka vs. RabbitMQ vs. SQS, delivery guarantees, backpressure.
- Microservices — service decomposition, REST vs. gRPC, API gateways, circuit breakers.
- API design — REST vs. GraphQL, versioning, rate limiting, auth, pagination.
- Load balancing and scaling — horizontal vs. vertical, stateless design, autoscaling.
- Networking and CDN — DNS, TCP/IP fundamentals, edge caching.
How to structure a system design answer
- Clarify requirements (2–3 min)
- Estimate scale (1–2 min)
- High-level design (5–10 min)
- Dive deep on bottlenecks (10–15 min)
- Handle trade-offs explicitly
Companies that test system design heavily
| Company | Roles | Typical depth |
|---|---|---|
| SRE, Cloud Eng, Backend | Very deep, distributed systems at scale | |
| Netflix | SRE, Backend, DevOps | Deep, reliability and fault tolerance |
| Microsoft | Backend, Cloud Eng | Mid-deep, Azure integration often comes up |
| TCS, Infosys, HCL | Backend, Full Stack | Mid, architecture reasoning and trade-offs |
| Oracle | Backend, DBA | Deep on database design specifically |