How it worksPricingInterview questions
Infosys · DevOps Engineer

Infosys DevOps Engineer Interview Questions (2026): process, questions and answers

Updated 4 November 2026 · prep guide

The Infosys DevOps Engineer interview usually probes your CI/CD knowledge, cloud native tooling and incident handling. Use this guide to rehearse each question, understand the reasoning behind answers and follow the last‑minute plan to boost confidence.

Preparing for Infosys? Rehearse these questions with Interview AI Helper: ask any question and get a suggested answer in seconds, written from your own resume.

Free 5-minute demo

How the Infosys DevOps Engineer interview usually works

  1. Application review: Candidates usually clear the HR screen and receive a brief on the technical interview format.
  2. Technical round: Often focuses on Linux, Docker, Kubernetes, Terraform and monitoring concepts.
  3. Scenario discussion: Typically involves a real‑world problem you might face on a client project.
  4. HR round: Usually assesses cultural fit and communication skills.
  5. Offer: If you meet the expectations, the recruiter shares the final offer details.

Technical questions and answers

What command shows the image layers of a Docker image and their sizes?

Use docker history --no-trunc <image>; it lists each layer, its size and the command that created it, helping you spot unnecessary layers and optimise the image.

How does a Kubernetes readiness probe differ from a liveness probe?

Readiness determines if a pod can receive traffic; failing it removes the pod from service endpoints but does not restart it. Liveness checks if the container is healthy; failure triggers a restart according to the pod's restart policy.

Explain the purpose of a Terraform state lock and how it works.

A state lock prevents concurrent runs from corrupting the state file. When a run starts, Terraform writes a lock record to the backend (for example, an S3 object with DynamoDB). Subsequent runs detect the lock and wait or abort until the first run finishes and releases the lock.

Which Linux command would you use to find processes consuming more than 80% CPU for the last minute?

Run top -b -n1 | awk '$9>80 {print $0}'; top in batch mode shows a snapshot, and the awk filter prints lines where the CPU column exceeds 80 percent.

Describe how you would roll back a failed Helm release.

First list releases with helm history <release> to identify the previous revision. Then execute helm rollback <release> <revision>; Helm restores the prior chart version and Kubernetes objects, allowing you to investigate the failure without manual kubectl commands.

Interview with Infosys coming up?

Practise these DevOps Engineer questions out loud and get suggested answers in real time, based on your own resume. Free 5-minute demo, no subscription.

Try the free demo

What is the function of a service mesh like Istio in a microservice environment?

A service mesh provides traffic management, mutual TLS encryption, observability and fault injection without modifying application code. It operates via sidecar proxies that intercept all inbound and outbound calls, giving operators fine‑grained control over routing and security.

How can you securely inject AWS credentials into a Kubernetes pod at runtime?

Create an IAM role with the needed permissions, then enable IRSA (IAM Roles for Service Accounts) and annotate the service account used by the pod. The pod receives temporary credentials via the AWS SDK without storing static keys.

What does the --no-cache flag do in a docker build command?

It disables the build cache, forcing Docker to re‑execute every instruction in the Dockerfile. Use it when you suspect cached layers hide changes, such as updated base images or package versions.

Explain how you would set up alerting for a sudden increase in error rate in a Prometheus‑monitored service.

Create a PromQL rule that calculates rate(http_requests_total{status=~"5.."}[5m]) and compare it to a threshold. Then configure Alertmanager to send a notification via email or Slack when the rule fires, ensuring the alert includes the service name and time window.

Why is it recommended to keep the number of containers per host limited, and how do you enforce it?

Limiting containers prevents resource contention and simplifies troubleshooting. Enforce it with Kubernetes ResourceQuota objects that cap CPU, memory and pod counts per namespace, or configure Docker daemon --max-concurrent-downloads and --max-concurrent-uploads for host‑level limits.

Scenario questions

A client’s CI pipeline fails when a new Helm chart version is deployed to a staging cluster because the service account lacks permissions. How would you resolve it?

Identify the missing RBAC rules by checking the pod’s events. Then create or update a Role/ClusterRole with the required get, list, watch on the resources the chart needs, and bind it to the service account using a RoleBinding. Verify the pipeline passes after applying the changes.

During a production incident, the monitoring system shows high latency spikes on a microservice, but logs are empty. What steps do you take?

First check the pod’s resource usage with kubectl top pod to rule out CPU or memory throttling. Next, verify network latency using kubectl exec and curl to dependent services. If the pod is healthy, inspect the service mesh or ingress for request throttling, and finally, enable debug logging temporarily to capture more details.

You need to migrate an on‑prem Terraform state to a remote backend without downtime. Outline your approach.

Export the current state file, then add a backend block for the remote store (for example, Azure Blob or GCS) in the configuration. Run terraform init -migrate-state to copy the state safely. After migration, verify that terraform plan shows no changes and that the remote backend is storing the state correctly.

HR and behavioural questions

Tell us about a time you had to convince a team to adopt a new CI/CD tool.

Use the STAR format: Situation, the team used manual scripts; Task, you needed faster feedback; Action, you built a proof‑of‑concept pipeline, demonstrated reduced build time and fewer errors, addressed concerns about learning curve with training sessions; Result, the team adopted the tool, cutting release cycle by 30 percent.

Describe a situation where you missed an incident deadline and what you learned.

Explain the context, the missed deadline, the root cause such as insufficient monitoring, the corrective actions like adding alert thresholds and post‑mortem documentation, and how you improved response times thereafter.

How do you handle conflicting priorities when multiple projects need your DevOps support simultaneously?

Prioritise based on business impact, communicate transparently with stakeholders, negotiate realistic timelines, and use automation to create reusable pipelines that serve multiple projects, reducing manual effort.

Give an example of when you automated a repetitive task and the benefit it brought.

Describe the manual task, the script or tool you built (e.g., a Bash script that rotates logs and archives them), and quantify the time saved or error reduction, highlighting the value to the team.

Your last 48 hours before the interview

  1. Review each technical question and rehearse the answer aloud
  2. Run a small Docker‑Compose project to practice docker history and docker build --no-cache
  3. Deploy a sample Helm chart on a local Kind cluster and perform a rollback
  4. Create a Prometheus alert rule for error rate and test it with alertmanager
  5. Write a concise STAR story for one behavioural question
  6. Mock a 30‑minute interview with a peer and request feedback

Common mistakes to avoid

Frequently asked questions

What is the typical duration of the Infosys DevOps interview process?

Candidates often report that the whole process takes a few weeks, with an initial HR screen, one or two technical rounds and a final HR discussion.

Do I need to know specific cloud platforms for the Infosys DevOps role?

While the company works across multiple clouds, interviewers usually focus on core concepts that apply to any platform, such as IaC, container orchestration and monitoring.

How important is hands‑on experience versus theoretical knowledge?

Interviewers commonly value practical experience; they prefer candidates who can demonstrate commands, scripts or real‑world scenarios rather than only definitions.

Will there be a coding test in the DevOps interview?

A small scripting exercise is often included to assess your ability to automate tasks, for example writing a Bash snippet that parses log files or creates a Terraform variable file.

Interview with Infosys coming up?

Practise these DevOps Engineer questions out loud and get suggested answers in real time, based on your own resume. Free 5-minute demo, no subscription.

Try the free demo

More Infosys interview guides

DevOps Engineer interview questions at other companies

Related reading

This is an independent preparation guide based on commonly reported interview patterns. The process varies by team, location and year. Interview AI Helper is not affiliated with Infosys.

← All interview question guides