Role Summary The DevOps/SRE Engineer builds and runs the infrastructure that makes the platform reliable, scalable and affordable. You own environments, delivery pipelines, GPU capacity and the observability needed to meet GA reliability targets.

Key Responsibilities

Infrastructure and Environments

- IaC: Define all cloud infrastructure as code (Terraform modules, remote state, policy checks with OPA/Checkov); separate accounts/subscriptions for dev, test, staging and production.
- Network: Design VPC/VNet layout, private endpoints, ingress with WAF, egress controls, service-to-service mTLS (service mesh or equivalent).
- Data services: Provision and operate managed MongoDB (Atlas or self-managed), Redis, Kafka/MSK, object storage with lifecycle rules, and backups.

Kubernetes and GPU

- Clusters: Operate EKS/AKS with node pools by workload (API, workers, GPU), pod security standards, resource quotas and namespaces per workplace.
- GPU pools: Run GPU node pools with the NVIDIA GPU Operator; use MIG or time-slicing where appropriate; use spot/preemptible capacity with fallbacks for batch jobs.
- Autoscaling: Implement queue-depth driven autoscaling (KEDA) and node provisioning (Karpenter or Cluster Autoscaler) with scale-to-zero for idle GPUs.

CI/CD and Release

- Pipelines: Build CI/CD for services, front end, models and infrastructure with test, scan (SAST, dependency, container, IaC) and signing gates; produce SBOMs.
- Progressive delivery: Implement blue-green and canary releases (Argo Rollouts or equivalent) with automated analysis and rollback; manage database migration safety.
- GitOps: Use Argo CD or Flux for declarative deployments and drift detection.

Observability and Reliability

- Telemetry: Implement metrics, logs and traces with OpenTelemetry, Prometheus,



Grafana and Loki (or Datadog/CloudWatch); tenant-aware dashboards.
- SLOs: Define SLIs/SLOs and error budgets for API, ingestion and inference; alert on burn rate; maintain on-call, runbooks and post-incident reviews.
- Resilience: Implement backups, disaster recovery (RPO/RTO targets) and regular restore tests with Velero or equivalent; support load and chaos testing before GA.

Cost and Governance

- FinOps: Tag resources and report cost by service and tenant (Kubecost, cloud cost tools); drive rightsizing, storage tiering and GPU utilisation improvements.
- Security partnership: Implement secrets management (Secrets Manager/Key Vault/Vault, external-secrets), patching and image hardening with the Security specialist.

Required Skills and Experience

- 5 years in DevOps, SRE or platform engineering.

• Production Kubernetes experience, including GPU workloads. • Infrastructure as code (Terraform preferred) on AWS or Azure.

• CI/CD pipeline design and release strategies (blue-green, canary).

- Prometheus/Grafana or equivalent observability stack; incident management.
- Linux, networking, TLS and IAM fundamentals; scripting in Bash or Python.

Preferred / Added Advantage

- NVIDIA GPU Operator, MIG, DCGM exporter.
- KEDA, Karpenter, Argo CD/Rollouts, service mesh.
- FinOps tooling and cost allocation.
- Managed Kafka and MongoDB operations.
- Cloud or CKA/CKAD certifications.

Key Deliverables

- Terraform modules and environment definitions.
- CI/CD pipelines with quality and security gates.
- GPU autoscaling design and validation.
- Observability dashboards, SLOs and alerts.
- DR plan with tested restore evidence; cost dashboards.

Education Bachelor’s degree in Computer Science, Engineering or equivalent experience.

Pay: ₹1,000,000.00 - ₹1,500,000.00 per year

Work Location: In person

📌 Devops/SRE Engineer (Hyderabad)
🏢 Spatial Alphabet
📍 Hyderabad