Divya Menon
Platform Engineer
Bengaluru, India | your.email@example.com | +91 98XXX XXXXX | linkedin.com/in/your-profile
Professional Summary
Platform Engineer with 4 years building internal developer platforms and Kubernetes infrastructure. At Cloudlink India, delivered a Backstage-based IDP that reduced infrastructure ticket volume by 75% across 60 engineers and cut pipeline failure rate from 12% to 1.8% by replacing a brittle Jenkins shared library with reusable GitHub Actions composite actions. Introduced Karpenter to production EKS, reducing node provisioning latency from 4 minutes to 45 seconds and cutting compute spend by 18%. Treats internal engineers as customers and platform adoption as the primary success metric.
Skills and Expertise
Certifications
- Certified Kubernetes Administrator (CKA)Mar 2024 - Mar 2027
- AWS Certified Solutions Architect - AssociateSep 2022 - Sep 2025
- HashiCorp Certified: Terraform AssociateNov 2022 - Nov 2025
Achievements
- Developer Productivity Award 2024 - Recognised for reducing average environment provisioning time from 3 days to 15 minutes via self-service Backstage templates.
Work Experience
- Designed and launched a Backstage-based open-source IDP serving 60 engineers, backed by Crossplane for infrastructure provisioning; self-service environment creation reduced infrastructure ticket volume from 20/week to under 5/week within 8 weeks of go-live, freeing the platform team from reactive support for the first time.
- Authored golden path Helm chart templates for microservices and batch workloads, reducing per-service boilerplate from 400 lines to 47 lines; all 8 product teams adopted the templates within 6 weeks, eliminating a source of cluster-wide misconfiguration.
- Manages 4 EKS clusters (dev, staging, production, shared-services) via Terraform and ArgoCD; migrated all cluster configuration to GitOps, reducing configuration drift incidents from 6 per quarter to zero and enabling full cluster state audit from Git history.
- Introduced Karpenter to replace Cluster Autoscaler on production EKS; node provisioning latency dropped from 4 minutes to 45 seconds, directly accelerating CI/CD feedback loops for all 8 product teams, while monthly compute spend fell by 18% through improved bin-packing and Spot Instance integration.
- Enforced OPA/Gatekeeper policies across all 4 clusters covering container resource limits, image registry restrictions, and namespace labelling; blocked 34 non-compliant deployments in the first month without requiring human review, all resolved by teams within 24 hours using automated policy violation messages.
- Replaced a shared Jenkins master with distributed GitHub Actions runners and hermetic build environments; pipeline failure rate dropped from 12% to 1.8% over 8 weeks and average build duration fell by 35%.
- Deployed centralised Prometheus/Grafana/Loki observability stack; onboarded 6 product teams with pre-built service dashboards, reducing mean time to detect production anomalies from 14 minutes to under 3 minutes across onboarded services.
- Maintained 2 EKS clusters and 15 ArgoCD applications; managed 4 cluster upgrades (v1.21 to v1.25) with zero unplanned downtime by following rolling node group upgrade procedures and pre-upgrade application compatibility checks.
- Migrated CI system from Jenkins to GitHub Actions for 12 repositories; average build time fell by 35% and the 2-person ops burden of maintaining the Jenkins master was eliminated entirely.
- Authored Terraform modules for VPC, EKS, RDS, and IAM that became the internal standard; 3 new projects launched using the modules without any infrastructure engineering involvement, reducing project bootstrap time from 2 weeks to 1 day.
- Introduced Helm chart versioning and a private OCI Helm registry on AWS ECR; replaced ad-hoc kubectl apply workflows across 4 teams, giving the platform team full visibility into deployed chart versions for the first time.
- Maintained Jenkins pipelines for 5 microservices; resolved an average of 4 build failures per week, keeping deployment SLA within agreed 4-hour window for all production releases.
- Automated EC2 start/stop scheduling for 12 non-production instances using Bash and cron, saving approximately INR 18,000/month in compute costs.
Education
Polishing your CV to land a cloud or DevOps role?
Optional 1:1 sessions with a senior cloud architect - CV review, interview coaching, and career direction. All site content stays free.
Was this helpful? Have a correction or suggestion?