Key Responsibilities

Cloud Infrastructure (AWS)

- Design, build and operate production AWS infrastructure: VPC/networking, IAM (least-privilege), EC2,

RDS/Aurora, S3, CloudFront, Route 53, ElastiCache, SQS/SNS, API Gateway.
- Infrastructure-as-Code everything — Terraform (preferred) or CloudFormation/CDK; no console-only changes in production.
- Multi-environment strategy (dev/staging/prod) with repeatable, auditable provisioning.

Containerization & Kubernetes

- Own our container platform: Docker, Amazon EKS, and ECS/Fargate — cluster architecture, node groups,

Karpenter/Cluster Autoscaler, Helm-based deployments, HPA/VPA, ingress and load balancing.
- Deploy, scale and troubleshoot microservices on Kubernetes; establish namespace/quotas/network policies for multi-team usage.
- Migrate and run stateful and latency-sensitive workloads (including AI inference services) on EKS.

AI/ML Platform Engineering & MLOps

- Operate and optimize AWS AI/ML workloads: SageMaker (training jobs, endpoints, pipelines, feature store)

and Amazon Bedrock (foundation model access, guardrails, provisioned throughput).
- Deploy and serve ML/LLM models on EKS/Kubernetes using tools such as KServe, Seldon Core, NVIDIA

Triton, or vLLM; manage GPU node pools, driver setup, and cost-effective GPU scheduling.
- Support RAG/vector-database infrastructure (pgvector, Pinecone, Qdrant), model versioning/registry,

MLflow-style experiment tracking, and inference autoscaling.
- Own the ML deployment pipeline: model build ® container ® canary deploy on EKS/SageMaker ® rollback.

CI/CD & Release Automation





- Build and maintain CI/CD pipelines (GitHub Actions, GitLab CI, Jenkins) with automated testing, image scanning and progressive delivery (blue-green/canary).
- Drive GitOps practices (ArgoCD/Flux) for declarative, auditable Kubernetes deployments.

DevSecOps

- Embed security across the SDLC: SAST/DAST (SonarQube, Trivy, Checkov, Snyk), container/registry image scanning, dependency and IaC scanning in pipelines.
- Manage secrets properly (AWS Secrets Manager/KMS, Vault), enforce IAM least privilege, and harden clusters to CIS benchmarks.
- Own AWS security posture: WAF, GuardDuty, Security Hub, CloudTrail audit logging; vulnerability remediation SLAs.
- Support compliance needs appropriate for a listed company (ISO 27001/SOC 2 posture), DR strategy,

automated backups and restore drills.

Observability & SRE

- Build full observability: CloudWatch, Prometheus/Grafana, centralized logging (OpenSearch/ELK), tracing and alerting.
- Define SLOs, run on-call with structured incident management and blameless RCAs; drive reliability improvements.

FinOps & Cost Optimization (Good to Have)

- AWS cost governance: tagging strategy, Cost Explorer/Budgets, CUR analysis; right-sizing, Spot Instances,

Graviton,



Savings Plans; showback reporting to engineering teams.
- Optimize the cost of AI/ML inference (GPU utilisation, batching, endpoint autoscaling) — a big lever for an

AI-first platform.

Must-Have Skills

- 5–9 years in DevOps/Platform/SRE roles, with 3+ years deep on AWS.
- Strong hands-on Kubernetes (EKS/ECS) in production — you can debug a CrashLoopBackOff and design a cluster from scratch.
- Terraform (or equivalent IaC) and Git-based workflow discipline.
- Solid Linux fundamentals, networking (VPC, DNS, TLS, load balancing) and scripting (Python, Bash).
- CI/CD pipeline design and implementation at scale; container image best practices.
- Hands-on DevSecOps: pipeline-integrated scanning, secrets management, IAM, and hardening.
- Experience running AI/ML workloads on AWS — SageMaker, Bedrock, and/or deploying ML models to

EKS/Kubernetes (GPU scheduling, model serving).
- Solid communicator who documents runbooks and works with product + AI teams without needing to be managed.

Good to Have

- FinOps experience — AWS cost dashboards, Savings Plans/Spot strategy, unit economics of LLM inference.
- KServe/Seldon/Triton/vLLM, Ray on Kubernetes, MLflow, feature stores.
- ArgoCD/Flux GitOps, Istio/service mesh, EKS Anywhere or hybrid cloud.
- EdTech/LMS/high-traffic consumer platform experience; experience in a listed-company/compliance environment.
- Certifications: AWS DevOps Engineer Professional, AWS Solutions Architect Associate/Professional, CKA,

Kubernetes Security Specialist (CKS).

📌 DevOps Engineer (AWS, Kubernetes & AI/ML Platform) (Mumbai)
🏢 Runtime Solutions
📍 Mumbai