Top 40 System Design Interview Questions | Must-Prepare before interview.

Top 40 System Design Interview Questions | Must-Prepare before interview Work through these 40 questions out loud. Open any question for the full model answer, the trap to avoid, and the follow-up a panel usually asks.

  1. @RestController vs @Controller?

    @Controller is for traditional MVC — return "user/list" view name. @RestController adds @ResponseBody so return types become JSON/XML body — standard for REST APIs. Use...

    Full answer

  2. How does @Transactional work?

    Spring creates a proxy around @Transactional beans. Calling createOrder() from another bean starts a transaction on the thread-bound EntityManager; commit on success, rollback...

    Full answer

  3. Explain JWT authentication flow in Spring Security.

    POST /login with credentials → auth provider validates → issue JWT with exp/sub/roles. Subsequent requests include Bearer token. JwtAuthenticationFilter runs before...

    Full answer

  4. Explain CAP theorem in the context of microservices.

    When a network split isolates nodes, you cannot have both perfect consistency and every node answering. Example: order service during partition — either reject writes (CP) or...

    Full answer

  5. What is the difference between ArrayList and LinkedList?

    ArrayList: dynamic array, great for get(i) and iteration, amortized append. LinkedList: node chain, costly random access, only wins for frequent insert/remove in middle with...

    Full answer

  6. How would you design a rate-limited REST API?

    API gateway or filter applies token bucket per API key — Redis stores tokens with TTL. Exceeding limit returns 429 + Retry-After. Different tiers per plan. Log rejections for...

    Full answer

  7. What is circuit breaker pattern?

    When payment service errors exceed 50% in 10s window, breaker opens — return fallback immediately for 30s. Half-open lets one request through to test recovery. Pair with...

    Full answer

  8. Explain microservices vs monolith trade-offs.

    Start monolith until team or scale pain justifies split. Microservices enable independent deploy and tech diversity but add network failures, observability burden, and data...

    Full answer

  9. What is an ExecutorService and why use it?

    Instead of new Thread per task, submit Runnable to fixed pool sized to CPU/IO profile. Handle Future results and exceptions. On shutdown hook call shutdown() and...

    Full answer

  10. What are virtual threads in Java 21 and when should you use them?

    Virtual threads excel when code blocks on IO — HTTP clients, JDBC, file reads — allowing one platform thread to serve thousands of concurrent requests. For CPU-heavy parallel...

    Full answer

  11. How do CompletableFuture and virtual threads compare for concurrency?

    Use virtual threads when business logic is naturally blocking — call DB, call HTTP, return. Use CompletableFuture when composing multiple independent async results with...

    Full answer

  12. What are weak references and WeakHashMap use cases?

    Attach metadata to objects without extending their lifetime — canonical mapping while object lives. When key GCs, entry drops. Do not rely on timing. For API response caching...

    Full answer

  13. What is Hibernate second-level cache and when enable it?

    Enable for ProductCategory, Currency — READ_ONLY, rarely changed. spring.jpa.properties.hibernate.cache.use_second_level_cache=true with Caffeine region factory. Avoid on...

    Full answer

  14. What are idempotency keys and how implement them in REST APIs?

    Client sends Idempotency-Key: uuid on POST /payments. Server Redis SETNX key, process, store response JSON. Retry with same key returns 201 + original body without double...

    Full answer

  15. Saga pattern — choreography vs orchestration?

    Order saga: reserve inventory → charge payment → ship. Failure at payment triggers release inventory compensation. Orchestration clearer for complex flows with visibility....

    Full answer

  16. Explain the transactional outbox pattern.

    OrderService @Transactional saves Order and OutboxEvent(type=ORDER_CREATED, payload=json). OutboxProcessor @Scheduled polls FOR UPDATE SKIP LOCKED, sends to Kafka, updates...

    Full answer

  17. When accept eventual consistency and how communicate it to clients?

    Checkout returns 202 Accepted with orderId — poll or webhook when confirmed. Inventory display may lag seconds — show "approximate stock." Payment ledger uses strong...

    Full answer

  18. What is the Backend-for-Frontend (BFF) pattern?

    Mobile BFF /mobile/home aggregates user profile + feed + notifications in one round trip. Web BFF different field selection. BFF handles OAuth BFF cookie pattern. Domain...

    Full answer

  19. Database sharding strategies for high-scale Java services?

    Shard by tenant_id mod 32 — RoutingDataSource lookup. User-scoped queries single shard. Global admin reports from CQRS read replica or Elasticsearch. Plan shard key early —...

    Full answer

  20. When and why use CQRS (Command Query Responsibility Segregation)?

    E-commerce: write Order aggregate to PostgreSQL; project OrderSummaryView to Elasticsearch for search/filter dashboard. Accept projection lag. Not for simple User CRUD. Combine...

    Full answer

  21. Load shedding and backpressure in microservices?

    Tomcat max queue + reject policy returns 503 with Retry-After. WebClient timeout + circuit breaker stops calling slow dependency. Kafka consumer pause partitions when...

    Full answer

  22. Kafka consumer groups, offsets, and delivery semantics?

    @KafkaListener with AckMode.MANUAL — ack after DB save. Consumer group order-service scales to partition count max. Idempotency key in message header dedup table. Monitor...

    Full answer

  23. Redis caching patterns: cache-aside vs write-through?

    @Cacheable("users") on findById — Redis cache-aside via Spring Data Redis. Update user: @CacheEvict. TTL 15m + jitter. Hot product page: lock on miss with Redisson. Do not...

    Full answer

  24. Your application memory usage keeps increasing over time. What do you suspect?

    Suspect memory leak if heap used grows after GC without traffic drop. Capture heap dump on OOM (`-XX:+HeapDumpOnOutOfMemoryError`), compare dominator tree. Common in Spring:...

    Full answer

  25. You deployed a new version but users still hit the old behavior. Why?

    Check K8s rollout status — old ReplicaSet still receiving traffic. Bust cache on static assets; API GET with `Cache-Control`. Confirm image tag in running pod matches CI build....

    Full answer

  26. API returns correct data but response time is inconsistent. Why?

    Tail latency often from stop-the-world GC, connection pool contention, or cold cache. Monitor p99 separately from mean. ```yaml spring.datasource.hikari.maximum-pool-size: 10...

    Full answer

  27. You enabled caching but performance did not improve. Why?

    Measure hits/misses. Wrong key expression caches per wrong granularity. ```java @Cacheable(value = "users", key = "#id") public User findById(Long id) { return...

    Full answer

  28. Autoscaling enabled but error rate increases. Why?

    New pods accept traffic before JVM/Caffeine warm. HPA adds API pods but Postgres max connections fixed. ```yaml readinessProbe: initialDelaySeconds: 30 ``` Scale connection...

    Full answer

  29. What is Anthropic prompt caching in LangChain4j 1.22.0, and when is it worth turning on?

    I put the interviewer persona, tool JSON, and the static policy appendix in the cached system prefix. The user question and this-request RAG chunks stay dynamic. If the prefix...

    Full answer

  30. Kubernetes liveness vs readiness vs startup probes?

    deployment.yaml: startupProbe HTTP /actuator/health with failureThreshold*period covering 60s startup. readinessProbe /actuator/health/readiness includes DB. livenessProbe...

    Full answer

  31. Your service needs zero downtime deployment. How do you achieve it?

    K8s rolling update + readiness probe ensures traffic only hits healthy pods. ```yaml readinessProbe: httpGet: path: /actuator/health/readiness port: 8080 initialDelaySeconds:...

    Full answer

  32. How do you trace a request across multiple microservices?

    Inject trace context at edge; log `traceId` in every service. ```java log.info("traceId={} orderId={}", MDC.get("traceId"), orderId); ``` Spring Boot 3:...

    Full answer

  33. Your Spring Boot app fails only when deployed in Docker. Why?

    Container `-Xmx256m` with 256Mi limit causes OOMKilled. Missing `DB_HOST` defaults to localhost inside container. ```dockerfile ENV JAVA_OPTS="-XX:MaxRAMPercentage=75.0" ```...

    Full answer

  34. Explain the difference between == and .equals() in Java.

    For primitives, == compares values. For objects, == asks if two references point to the same instance. equals() compares content when overridden — String overrides it for...

    Full answer

  35. How does Spring Boot auto-configuration work?

    @SpringBootApplication enables @EnableAutoConfiguration, which loads candidate configs from starters. Each auto-config uses @ConditionalOn* to register beans only when...

    Full answer

  36. Describe the Spring Bean lifecycle.

    Container instantiates the bean, injects dependencies, runs BeanPostProcessors, then @PostConstruct or InitializingBean. Singleton lives for context lifetime; prototype creates...

    Full answer

  37. What is the N+1 query problem and how do you fix it?

    Loading 100 orders then accessing each order's items triggers 101 queries. Fix with JOIN FETCH in JPQL, @EntityGraph on repository method, or a DTO query selecting only...

    Full answer

  38. HashMap vs ConcurrentHashMap — when and why?

    HashMap for single-threaded or external synchronization. ConcurrentHashMap for shared concurrent maps — finer-grained locking than Hashtable. Cannot store null keys/values in...

    Full answer

  39. Explain immutability and why String is immutable.

    Immutable objects cannot change after build — thread-safe without synchronization. String immutability allows intern pool, caching hashCode, and security when used as...

    Full answer

  40. What is dependency injection and why does Spring use it?

    DI supplies collaborators from outside the class — typically via constructor. Problem solved: tight coupling to concrete classes, hard unit tests, scattered object...

    Full answer

System Design Interview Questions and Answers · Top 50 System Design Interview Questions and Answers (2026 Edition) · System Design Interview Questions Every Fresher Must Know · System Design Interview Questions for 2–3 Years

All interview question lists · Interview guides · Home · Support the platform