Sr Software Engineer - Site Reliability Engineering
Summary
Senior SRE at Asurion's India capability centre in Pune, owning SLOs, observability, incident response, toil reduction and resilience, plus platform/DevOps work: AWS (EKS, ECS, Lambda), Terraform, GitHub Actions/GitOps CI/CD and PostgreSQL/MySQL operations, with AI-assisted ops via LangGraph/Bedrock Agents. Needs 3-5 years SRE/DevOps experience and production Python/JavaScript.
What wearelookingfor:
Experienceandeducation:
- Bachelor's degree in Computer Science, InformationTechnology, Electronics or a related engineering discipline.Amaster's (ME, MTech, MS or MCA) in a computing or quantitative discipline is preferred.
- Threetofiveyearsofprofessionalsoftwareengineering,withademonstrablerecordofshipping andsupportingproductionsystems.
- Stronghands-on SRE,DevOpsorplatformengineering,includingcarryingapagerforaproductionenvironmentofrealconsequence.
- Substantial Kubernetes andTerraform experience inproduction rather than in alab, and a significant componentowned end to end.
- Significant incidents handled, with credible post-incident practice, and production-quality code in Python or JavaScript rather than only shell glue.
- Cloudcostoptimisationwithquantifiedresults, orout-of-hourssupportmodelexperience,carriesrealweighthere.
Coretechnical skills:
- AWS:EKS, ECS,EC2, Lambda, VPCand networking,IAM, KMS, Route53, CloudFront, S3,CloudWatch,Terraformand BedrockAgents.
- GitHub and GitOps:GitHub Actions as the delivery platform, with reusable workflows, environment protection rules and OIDC federation toAWS instead of long-lived keys; pull-request based change with branch protection, required reviews and Terraform plans reviewed before apply; GitHubAdvanced Security checks in the pipeline; and GitOps practice where Git is the source of truth for infrastructure and deployments, withdrift detected and reconciled.
- PythonandJavaScript:production-standardcodeforautomation andtooling,notonlyshellglue.
- Open Telemetry, Dynatrace and CloudWatch:instrumentation and dashboards that get used, with OpenSearch for log and trace analytics and Grafana as an alternative, plus service-level objective design and toil measurement.
- PostgreSQLandMySQL:RDSorAuroraoperations,querybehaviourunderload,failoverandbackupverification.
Nice to have:
- Databricks platform operations, since the regional data platform runs there.
- FinOpspractice,commitmentplanningandquantifiedsavings.
- Vulnerability remediation and secrets management at platform level.
- AWS DevOps Engineer Professional or SolutionsArchitect certification.
AI-assistedoperations:
- Tooling writtenagainst amodelAPI,and anagent frameworksuch asLangGraph orBedrockAgentsused whereitremoves realtoil.
- Model Context Protocol integrations that let an assistant read internal systems under scoped, auditable permissions, with a human decision in front of anything that changes production.
How you work:
- Owns outcomes, not tickets.Takes a vague requirement through to something running in production.
- Raises the bar around them. Reviews carefully, and mentors the engineers nearby.
- Calm and effective in an incident, and honest afterwards about the cause.
- Clear in writing. Design notes, runbooks and handover documentation others can follow.
- Pushes worktotherightownerratherthanquietlyabsorbing it.
Whatyouwillown:
Reliability engineering
- Service-level objectives:agreed with each engineering team, with error budgets used in real prioritisation conversations instead of as decoration.
- Observability:metrics, logs, traces and synthetic checks instrumented consistently enough that an engineer can diagnose an unfamiliar service.
- Alertingthatleadstoaction:removethealertsthatwakepeoplewithoutcause.Itisengineeringworkinitsownright.
- Incidentresponse: takeyourturnintherotation,anddiagnose,mitigate, communicateandcomplete thefollow-upactions.
- Toil, attackedsystematically:measure it,targetthe largestsourcesandautomate them.
- Resilience: capacity modelling,load testing,autoscaling behaviour,graceful degradation, anddisaster recoverythat getsexercised.
PlatformandDevOpsengineering:
- CI/CD pipelines:build,test,artefactmanagement,environmentpromotion,andprogressivedelivery withcanaryorblue-greendeployment.
- Infrastructure ascode:reusableTerraformmodules, stateanddriftmanagement,anda reviewstandard forinfrastructurechange.
- Thecontainerplatform: EKSclusterlifecycle, workloadstandards,autoscaling,ingress,RBACandupgrades.
- Developer self-service:engineeringteamsshouldbeableto provision,deployandobserve withoutwaitingonthis team.
- Secure defaults:secrets management, least-privilege IAM, network segmentation and patching automation, alongside the security engineering team.
- Cost engineering:visibilityandattribution,rightsizing,wasteeliminationandquantifiedsavings.
Productionoperations:
- Amulti-regionAPAC footprint:traffic peaks follow local business hours and local retail and operator events, so capacity, failover and maintenance windows are planned per region.Awindow that is quiet in one market is peak in another.
- Operationalreadiness:noserviceentersproductionwithoutrunbooks,alerts,dashboards,anamedownerandarollbackpath.
- Routineoperations:handleoperationaltasksandfirst-line response,andautomatethemoutofexistence wherepossible.
Craft,reviewandmentoring:
- Designbeforebuilding:short,cleardesignnotesfornon-trivialwork,takentoreview,withthetrade-offswrittendown.
- Review asrealwork:itshould showinhowyourweekis allocated.
- Leave theplatformbetter:refactorasyougo,removedeadcode,andimprovethetoolinganddocumentationyoutouch.
- Mentoring:pairwithlessexperiencedengineers, givereviewfeedback thatteaches,andhandoverworkthatstretchesthem.
- Documentedsystemknowledge:raisedocumentationandtestcoverage asyoutakeplatformson,soanewjoinerisproductive quickly.
AI-accelerateddevelopmentiscoretohowwework:
Every engineer here uses an approved assistant such as Claude Code, Cursor, GitHub Copilot or Databricks Assistant as part of normal work, and the expectation runs past autocomplete: a scoped multi-file task, a diff you review properly, and tests and static analysis as the gate.
Treat a prompt as an egress path. No customer data, no credentials, and nothing outside theAsurion acceptable-use and data-classification standards.
Skills
- AI
- Analytics
- Automation
- AWS
- Bash
- CI/CD
- Claude Code
- CloudFront
- CloudWatch
- Databricks
- DevOps
- Dynatrace
- EC2
- ECS
- EKS
- Git
- GitHub
- GitHub Actions
- Github Copilot
- GitOps
- Grafana
- IAM
- JavaScript
- Kubernetes
- Lambda
- Networking
- Observability
- OpenID
- OpenSearch
- OpenTelemetry
- Python
- Route 53
- S3
- Secrets Management
- SRE
- Terraform
Sign in to read and join the discussion.