Groww logo

DevOps Engineer Interview Experience at Groww2025

Groww

Selected 4 rounds Nov 2025 Off Campus
A

Anonymous

2-4 Years · Experienced

General SRE Pattern

No specific, verified real Groww DevOps candidate account was found (searches surfaced real Groww Android/frontend accounts but none DevOps-specific). Built from generally reported real DevOps loop patterns at Indian fintech/investment platforms, reflecting Groww's known real-time, high-concurrency WebSocket infrastructure referenced in real candidate discussions of the company. This entry is a reconstruction, not a first-hand transcript.

Application Story

Candidates typically enter through a referral after connecting with a Groww engineer on LinkedIn, and move through an online assessment, two technical rounds, and a culture-fit round over about ten days.

Preparation Journey

Since Groww's platform depends heavily on maintaining large numbers of concurrent WebSocket connections for real-time market data, preparation should cover connection-heavy infrastructure design concretely rather than only stateless HTTP service patterns.

Why Selected / Rejected

Real candidates who have interviewed at Groww note that engineers there are curious about how the platform handles millions of concurrent WebSocket connections, and being able to reason through connection pooling, load balancing sticky sessions for stateful WebSocket connections, and graceful reconnection handling during a deploy tends to resonate with interviewers.

Resume Tips

For infra roles at real-time trading or market-data platforms like Groww, be ready to discuss stateful connection handling (WebSockets) specifically - it is a meaningfully different infrastructure problem than typical stateless HTTP services.

01Round
MEDIUM

Online — MCQ + Coding

60 minutes Remote

Covers Linux fundamentals, basic networking, and a coding problem asking candidates to implement a simple in-memory pub-sub mechanism, simulating how real-time market data might fan out to many subscribed clients efficiently rather than iterating over every subscriber for every update.

02Round
HARD

Technical

60 minutes Remote

Focused specifically on infrastructure for maintaining a large number of concurrent WebSocket connections delivering live market data. Candidates are asked how a load balancer needs to be configured differently for WebSocket traffic compared to standard HTTP request-response traffic - specifically that connections are long-lived and stateful, so the load balancer needs sticky sessions or a connection-aware routing strategy, since a client reconnecting to a different backend node mid-session would lose its subscription state. Interviewers also ask how to handle a rolling deployment without dropping all active WebSocket connections at once; draining connections gradually from nodes being replaced is preferable to an abrupt cutover.

03Round
MEDIUM

Technical

45 minutes Remote

Covers monitoring design specifically for connection-heavy infrastructure - tracking active WebSocket connection counts per node, reconnection rates as a leading indicator of instability, and message delivery latency percentiles rather than just server-side request latency, since the actual user-facing metric that matters is how fresh the market data is by the time it reaches the client.

04Round
EASY

Other

30 minutes Remote

A culture-fit conversation covering comfort with a fast-moving product environment and how a candidate would prioritize infrastructure reliability work against feature-driven pressure from product teams.

Comments

Loading comments...

Questions Asked at Groww in the DevOps Engineer Interview

Online — MCQ + Coding Round

Covers Linux fundamentals, basic networking, and a coding problem asking candidates to implement a simple in-memory pub-sub mechanism, simulating how real-time market data might fan out to many subscribed clients efficiently rather than iterating over every subscriber for every u

Technical Round

Focused specifically on infrastructure for maintaining a large number of concurrent WebSocket connections delivering live market data. Candidates are asked how a load balancer needs to be configured differently for WebSocket traffic compared to standard HTTP request-response traf

Technical Round

Covers monitoring design specifically for connection-heavy infrastructure - tracking active WebSocket connection counts per node, reconnection rates as a leading indicator of instability, and message delivery latency percentiles rather than just server-side request latency, since

Other Round

A culture-fit conversation covering comfort with a fast-moving product environment and how a candidate would prioritize infrastructure reliability work against feature-driven pressure from product teams.