Nothing can limit your
dreams when you are at
Zoft

Home / Careers

Junior DevOps Engineer

Work Location : Kochi, Kerala

1. Role at a Glance


  • Position: Junior DevOps Engineer
  • Full Time Permanent
  • Experience: 0 to 2 years. Graduates are welcome: knowledge, certification and aptitude count for more than years
  • Location: Kochi, Kerala. On-site for the first six months, hybrid thereafter
  • Hours: 11:30 to 20:30 IST, Monday to Friday, with a daily overlap with the client's Eastern Time business hours. Production on-call rotaonce signed off
  • Reports to: Senior DevOps Engineer, for day-to-day direction, work review and development. Project Manager of each assigned project,for delivery. Delivery Manager, for line management, performance review and escalation
  • Works with: Project and programme teams (Technical Leads, Architects, QA, Production Support), and the client's DevOps, security and infrastructure engineers once signed off.
  • Start: Immediate. Candidates able to join within 30 days are preferred
  • First-year goal: Run the day-to-day DevOps work of the assigned projects independently: releases, on-call, routine infrastructure changes and client updates


2. The Role


The Junior DevOps Engineer keeps live Azure systems released safely and running reliably for our clients, working under the direction of a Senior DevOps Engineer. The role starts with learning the environment and carrying out routine tasks under review, and grows, task by task, until the engineer runs the day-to-day DevOps work of the assigned projects independently within the first year.


AI is central to how this role works. You will use AI every day to write, check and explain pipelines, infrastructure code, scripts and queries; to read telemetry and triage alerts; to draft release notes, change records and runbooks; and to learn the environment faster. You will also be expected to verify everything it produces before it is used.


We do not require prior job experience. We require evidence that the candidate already knows the fundamentals well, has proved it through recognised certifications and hands-on projects, uses AI well and safely, learns fast, communicates clearly, and can be relied on.



3. The Environment


What the work involves


  • Many live production systems at once: Programmes of integrated applications and shared platform services, including AI-powered processing, run in production for enterprise clients, each with its own release cadence, service levels and change process.
  • Integration at every step: A single business transaction can cross Azure API Management, Service Bus, Event Grid, Function Apps, Container Apps, AI services and the client's own systems.
  • Client-governed estates: The work runs inside client-owned Azure tenants and Azure DevOps organisations, under zero-public-access network policies and formal change advisory boards.
  • Direct contact with the client: Our clients are based in Canada and have their own DevOps, security and infrastructure engineers. This role speaks with them from the early months, with a daily overlap with their Eastern Time business hours.
  • Release control enforced by the pipeline: Build once, promote the same artefact through every environment, verify on the running system, and roll back automatically when a check fails.


4. What the role offers


  • A structured first year, from learning the environment to running it day to day, with a senior engineer directing your development.
  • A working practice built around AI-assisted DevOps, with company-certified tools and clear rules for verifying AI output.
  • Hands-on work across the full DevOps and operations range: pipelines, infrastructure as code, containers, observability, incident response and security.
  • A certification path aligned to the role, through to Microsoft Certified: DevOps Engineer Expert.
  • Direct exposure to a North American client's engineering teams.



5. Must-Haves


Each item is a requirement. A candidate must meet every one. Job experience is not required.


  1. Education and certification: Bachelor of Engineering or Technology in Computer Science, Information Technology, Electronics or an equivalent discipline. And at least one current certification from: Microsoft Certified: Azure Administrator Associate (AZ-104), or HashiCorp Certified: Terraform Associate.
  2. AI-assisted DevOps: Has used AI assistants to write, debug and explain pipeline YAML, Terraform, scripts or queries; checks the output by running and testing it; and can show a specific case where they found and corrected an AI error.
  3. Azure fundamentals, known well: Compute (App Service, Function Apps, Container Apps), networking (virtual networks, subnets, network security groups, DNS, private endpoints), identity (Microsoft Entra ID, RBAC, managed identities), Key Vault, Storage, and Azure Monitor.
  4. Linux, networking, scripting and Git: Linux command line; Bash, plus PowerShell or Python; HTTP, TLS and DNS basics; Git branching, pull requests and resolving merge conflicts.
  5. CI/CD, infrastructure as code and containers, built hands-on: Has built at least one working CI/CD pipeline (Azure DevOps or GitHub Actions), provisioned cloud resources with Terraform, and built and run a container with Docker, in an academic, personal or internship project whose code can be shown.
  6. Aptitude and learning speed: Strong logical reasoning and problem decomposition; learns an unfamiliar tool from its documentation and applies it within hours. Assessed at selection.
  7. Communication and etiquette: Clear, professional spoken and written English; can explain a technical problem simply, write a concise status update, and conduct themselves courteously and punctually in meetings with clients.
  8. Availability and focus: Able to work the stated hours, including the daily Eastern Time overlap; to work on-site in Kochi for the first six months; to join the on-call rota once signed off; and to sustain focused study alongside delivery work.



This Role Is Not for You If


  • You want to be told exactly what to do every day. This role expects you to take on more of the work, task by task, as you prove you can do it.
  • You hide a mistake or guess in production. Here, you stop, say so at once, and ask.
  • You avoid AI tools, or you trust them without checking. This role uses AI every day and verifies every output.
  • You learn only when a task forces you to. This role expects steady, self-driven study alongside delivery.
  • You find documentation tedious. Every task you perform ends with a runbook someone else can follow.
  • You are uncomfortable speaking to client engineers. Here, you speak with them from the early months and lead routine meetings within the first year.
  • You cannot keep the daily Eastern Time overlap, or join the on-call rota once signed off.



6. Key Accountabilities


You are accountable for the following on the projects assigned to you. In any task area not yet signed off (Section 7), you do the work and a senior engineer reviews it before it is applied.


6.1 AI-Assisted DevOps


  • Use company-certified AI tools every day to draft pipeline YAML, Terraform, scripts and KQL queries from a written specification. Run, test and be able to explain every line before raising a pull request, and mark AI-assisted changes as such.
  • Use AI to read telemetry: summarise alerts and logs, draft KQL to locate a fault, compare behaviour before and after a release, and propose likely causes for a reviewer to confirm.
  • Use AI in incident triage to find the right runbook and draft the status update. AI tools never execute an action on production.
  • Generate release notes, change records and runbook drafts from release manifests, pipeline output and your own notes, then edit them for accuracy.
  • Build and maintain the team's prompts, project rules and reusable AI templates for recurring DevOps tasks, so that AI output follows the programme's standards by default.
  • Support AI services running in production: monitor model endpoints, quotas, latency and cost, and release prompt and configuration changes through the pipeline like any other change.
  • Record each AI-assisted improvement with its measured effect, such as time saved or an error caught, and keep client secrets, credentials and data out of every prompt.


6.2 Release Engineering and CI/CD


  • Run pipelines and production releases: build once, promote by image digest, verify on the running system, and roll back automatically when a check fails.
  • Maintain pipeline templates and environment configuration through pull requests.
  • Keep the release record, notifications and Jira transitions complete for every release you run.
  • Stop a release and roll back the moment a check fails or a signal is unclear.


6.3 Production Operations and Observability


  • Watch dashboards and alerts for the systems in scope, triage alerts with their runbooks, and escalate at once when a runbook does not cover the situation.
  • Take your place on the on-call rota once signed off.
  • Run the operational calendar: patching, certificate and secret rotation, runtime end-of-support tracking, and scheduled restore tests.
  • Prepare the monthly service report for each system in scope.


6.4 Infrastructure as Code and Security


  • Make every infrastructure change through Terraform and a reviewed plan. No changes in the Azure portal, in any environment above Dev.
  • Run the weekly drift check and raise every drift as a defect.
  • Keep secrets in Key Vault with managed identities: never in code, pipeline variables, chat or prompts.
  • Request and use production access only through just-in-time elevation, and only for the task in hand.


6.5 Test Automation and Quality Gates


  • Maintain the smoke-suite infrastructure and the test environments, working with QA.
  • Provide synthetic test data on request. No real client data outside production.
  • Keep static-analysis and security quality gates running, and their results visible.


6.6 Client Communication


  • Join client technical meetings, take the notes, and send the written follow-up of decisions and actions the same working day.
  • Present routine status and release updates to the client's engineers, and lead routine technical meetings once signed off.
  • Be punctual, prepared, courteous and concise in every meeting and message. Refer any scope, cost or date question to the Project Manager.


6.7 Learning, Documentation and Reporting


  • Follow the certification plan: Microsoft Certified: Azure Administrator Associate (AZ-104) and HashiCorp Certified: Terraform Associate, whichever is not already held, within the first six months; Microsoft Certified: DevOps Engineer Expert (AZ-400) within the first year.
  • Keep a weekly learning log: what was learned, what was applied, and what is next.
  • Write or update a runbook for every operational task you perform, so that someone else can perform it from the runbook alone.
  • Contribute your section to the weekly DevOps report.
  • Report your own mistakes immediately, with what happened and what you have done about it.



7. Authority and Decision Rights


Independence is granted one task area at a time, as competence is shown, not by date.


  • Sign-off: a task area (for example running a release, making a Terraform change, rotating a certificate, taking primary on-call, or leading a routine client meeting) is signed off when you have performed it correctly under review several times, written or updated its runbook, and explained it back to your reviewer. The Senior DevOps Engineer records each sign-off.
  • Before sign-off: you do the work, and a senior engineer reviews it before it is applied, or is present while you do it.
  • After sign-off: you do it independently and report it.
  • Always: you may stop a release or trigger rollback at any time, without approval, and you escalate at once anything new, unclear or outside a runbook.
  • Never: waive a failed gate, change production outside the pipeline, let an AI tool act on production, or make scope, cost or date commitments to the client.



8. The First Year



By 3 Month:

What you do independently

Routine pipeline runs, drift checks, operational calendar tasks, meeting notes and follow-ups, runbooks, and AI-drafted queries and scripts that pass review.

What you do with a senior engineer

Production releases and on-call(shadowing); every production change


By 6 Month:

What you do independently

Routine production releases, secondary on-call, standard Terraform changes, and routine

client updates. AZ-104 and Terraform Associate held.

What you do with a senior engineer

Non-routine releases, primary on-call, and incident response


By 12 Month:

What you do independently

Day-to-day releases, primary on-call, routine infrastructure changes, the monthly service

report, and routine client technical meetings for the assigned projects. AZ-400 held

What you do with a senior engineer

New designs, major incidents, and changes that span projects



9. How Success Is Measured


  • Sign-offs: On the dates agreed in your development plan
  • Certifications: AZ-104 and Terraform Associate within 6 months; AZ-400 within 12 months
  • Production changes outside the pipeline: Zero
  • Actions on production executed by an AI tool: Zero
  • AI-assisted changes verified before merge and marked in the pull request: 100%
  • AI-assisted improvements recorded, with measured effect: At least one a month
  • Own mistakes reported: Every one, within the hour, with the facts
  • Runbooks: One for every operational task performed
  • Pull requests approved within one review round: 80% or more by month 9
  • On-call shifts covered as rostered: 100%
  • Client meeting follow-ups: Same working day, every meeting
  • Weekly learning log: Every week
  • Independence at 12 months: Day-to-day releases and on-call for the assigned projects run independently for four consecutive weeks



10. Expected Competencies


Assessed and weighted at interview.


  • AI practice: writes a precise specification before prompting; gives the tool the right context; checks the output against documentation; and recognises common AI errors in DevOps work, such as invented resource arguments, outdated API versions, insecure defaults and plausible but wrong KQL.
  • AI in operations: understands anomaly detection, alert correlation and dynamic thresholds in Azure Monitor, and where they mislead.
  • Responsible AI use: data protection, no secrets in prompts, human approval for every production action, and a record of AI-assisted changes.
  • Troubleshooting method: forms a hypothesis, tests it, narrows down, and stops to ask before acting in production.
  • Security habits: least privilege, secrets hygiene, the OWASP Top 10 at awareness level, and why production data never leaves production.
  • Query skills: basic SQL, and Kusto Query Language (KQL) at awareness level.
  • Operations awareness: what incidents, problems, changes and service levels are (ITIL 4 at awareness level), and why releases are rolled back rather than fixed forward.
  • Written English and etiquette: a status update or incident note that is short, factual and needs no rewrite; punctual, prepared and courteous, and respectful of the client's time and process.



11. Desirable


Considered when candidates are otherwise equal.


  • A current Microsoft AI certification at fundamentals or associate level.
  • A small AI-assisted tool built for a DevOps task, such as a log summariser, a pull-request checker or a runbook assistant, with the code available.
  • Microsoft Certified: Azure Fundamentals (AZ-900); Kubernetes and Cloud Native Associate (KCNA); GitHub Foundations or GitHub Actions; a Linux certification (LFCS or RHCSA); ITIL 4 Foundation.
  • An internship in cloud, DevOps or infrastructure.
  • Open-source contributions, a technical blog, a home lab, or hackathon work that can be shown.



12. Personal Attributes


  • Dependable: does what was agreed, when it was agreed, and says early when that will not happen.
  • Fast learner: closes knowledge gaps on their own initiative and applies what they learn the same week.
  • Sceptical of AI: uses it constantly, and trusts its output only after checking it.
  • Focused: finishes what they start, and works through problems without needing constant direction.
  • Honest: reports mistakes and uncertainty immediately, without being asked.
  • Courteous: professional in tone and conduct with colleagues and clients alike.
  • Composed: stays calm and methodical on an incident call.



13. Working Conditions and Compliance


  • Hours: 11:30 to 20:30 IST, Monday to Friday, giving a daily overlap with the client's Eastern Time business morning. Hours move on scheduled release days and during incidents.
  • On-call: shadowing first; on the rota once signed off, usually within the first six months, with out-of-hours response to incidents when required.
  • Location: on-site in Kochi for the first six months, so that learning happens alongside the team; hybrid thereafter.
  • Confidentiality: the role works on client engagements under NDA. Client names, systems, data and documents stay within the engagement.
  • Security: background verification before start; adherence to ZOFTSOLUTIONS information-security policies and the client's security requirements.
  • AI use: company-certified AI tools only; no client secrets, credentials or data in prompts; no AI tool acts on production.
  • Equal opportunity: ZOFTSOLUTIONS is an equal opportunity employer, and selection is made on the requirements in this document. Candidates who need an adjustment to the selection process may ask HR.



14. Selection


Shortlisted candidates complete an aptitude assessment, a hands-on cloud lab in which AI tools are allowed, an AI verification exercise, a learning-speed exercise, a communication exercise, and an interview.





Total Number of Openings : 2

Responsibilities

.

Skills

.

Work Experience

0-2 Years

Education

Bachelor of Engineering or Related

Work Mode

Hybrid

Interested? Apply now!