About Backend Developer interviews in India
Indian tech interviews typically run across multiple rounds — a technical screen, one or two deep-dive rounds, a system/design or practical round, and an HR and managerial discussion. Interviewers care as much about how you reason as the final answer, so think aloud, state assumptions, and use real examples from your own work.
🎯 Interview Success Tips
STAR MethodSituation → Task → Action → Result. Use for every behavioural question. Quantify the Result.
Research FirstRead company news, LinkedIn page, Glassdoor reviews and the interviewer's profile before the interview.
Salary TipNever give a number first. Ask: "What is the budgeted range for this role?" — always.
Virtual InterviewsTest camera + mic 30 min before. Good lighting, neutral background. Join 5 min early.
🔧 Technical Questions
Technical Question 1
What is the difference between a process and a thread?
💡 How to answer: Processes have isolated memory; threads share memory within a process and are lighter to create. Threads need synchronisation (locks) to avoid race conditions. Mention use cases — multiprocessing for CPU-bound, threads/async for I/O-bound.
Technical Question 2
Explain SQL vs NoSQL and when you'd choose each.
💡 How to answer: SQL: relational, ACID, fixed schema — best for transactions and complex joins. NoSQL: flexible schema, horizontal scale — best for high-volume, denormalised or rapidly-evolving data. Justify with a concrete scenario.
Technical Question 3
What makes an API RESTful? How is it different from GraphQL?
💡 How to answer: REST: stateless, resource URLs, HTTP verbs, status codes. GraphQL: single endpoint, client specifies exactly the fields it needs, avoiding over/under-fetching. Discuss caching and versioning tradeoffs.
Technical Question 4
How do you design a database schema and decide on indexing?
💡 How to answer: Model entities and relationships, normalise to remove redundancy, then selectively denormalise for read performance. Index columns used in WHERE/JOIN/ORDER BY, but remember indexes slow writes.
Technical Question 5
How do you optimise a slow query?
💡 How to answer: Run EXPLAIN/ANALYZE, add or fix indexes, avoid SELECT *, reduce N+1 queries, paginate, cache hot results in Redis, and consider read replicas or denormalisation for heavy reads.
Technical Question 6
What is the CAP theorem and how does it affect system design?
💡 How to answer: In a distributed system you can guarantee only two of Consistency, Availability and Partition tolerance. Since partitions happen, you trade C vs A. Give examples — banking favours C, social feeds favour A.
Technical Question 7
How do you handle authentication and authorisation in a backend?
💡 How to answer: Use proven mechanisms — JWT or session tokens, OAuth2 for delegation. Hash passwords with bcrypt/argon2, enforce RBAC, validate on every request, and never trust client input.
Technical Question 8
Explain caching strategies and cache invalidation.
💡 How to answer: Cache-aside, write-through, write-behind. Use Redis/Memcached for hot data, set TTLs, and version keys. Invalidation is hard — discuss stale-data tradeoffs and cache stampede protection.
Technical Question 9
How would you design a system to handle 1 million requests/day?
💡 How to answer: Clarify peak vs average, add a load balancer, scale stateless services horizontally, cache reads, use a queue for async work, choose the right DB with replicas, and add monitoring. Show capacity estimation.
Technical Question 10
What is idempotency and why does it matter in APIs?
💡 How to answer: An idempotent request produces the same result no matter how many times it's sent — vital for payments and retries. Implement with idempotency keys so a retried POST doesn't double-charge.
Technical Question 11
Explain message queues and when to use them.
💡 How to answer: Queues (RabbitMQ, Kafka, SQS) decouple producers from consumers, smooth traffic spikes, and enable async processing and retries. Use for emails, notifications, and event-driven pipelines.
🧠 Behavioural Questions
Behavioural Question 1
Describe a performance problem you solved on the backend.
💡 How to answer: Use STAR. Cover how you measured (profiling, slow-query logs), the bottleneck found, the fix, and the quantified improvement (latency down X%, cost saved).
Behavioural Question 2
Tell me about a time you made a wrong architectural decision.
💡 How to answer: Own it honestly, explain the context, what went wrong, how you migrated out of it, and the lesson. Shows maturity and reflection.
💡 Situational Questions
Situational Question 1
Your service is returning 500 errors in production. Walk me through your response.
💡 How to answer: Check dashboards and logs, identify the failing dependency, roll back or feature-flag the change, mitigate (rate-limit, scale), then root-cause and add a regression test and alert.
Situational Question 2
A third-party API you depend on is down. How do you keep your service resilient?
💡 How to answer: Use timeouts, retries with backoff, circuit breakers, fallbacks/cached responses, and graceful degradation. Queue work for later and alert. Never let one dependency take down everything.
Situational Question 3
You need to ship a feature but it requires a risky DB migration. How do you proceed?
💡 How to answer: Use a backward-compatible, multi-step migration (expand-migrate-contract), run it off-peak, test on a replica, keep a rollback plan, and avoid locking large tables.
💰 Salary Questions
Salary Question 1
What is your expected CTC as a backend developer?
💡 How to answer: Anchor on market: fresher ₹5–10 LPA, mid ₹12–24 LPA, senior ₹28 LPA+ in India by city and company tier. Ask for the band first.
Salary Question 2
We can offer a modest hike only. Can you reconsider?
💡 How to answer: Quantify impact (scalability, reliability you bring), cite AmbitionBox/Levels data, and if base is capped, negotiate ESOPs, joining bonus, or a 6-month review.
🎤 Ask Interviewer Questions
Ask Interviewer Question 1
What does the backend architecture look like — monolith or microservices?
💡 How to answer: Reveals tech debt, deployment complexity, and what you'll actually work on.
Ask Interviewer Question 2
How do you handle on-call and production incidents?
💡 How to answer: Shows operational maturity and work-life balance expectations.
Ask Interviewer Question 3
What are the biggest scaling challenges the team faces right now?
💡 How to answer: Signals you want to contribute to real problems, and reveals system health.