I keep a six-cluster production Kubernetes fleet at 99.9% uptime.
Senior SRE at Wizeline, embedded with Capsule — a US healthcare and pharmacy platform on AWS. Twelve years building software, five of them running cloud infrastructure: EKS, Terraform, Helm, Prometheus, and a lot of careful, zero-downtime migrations.
- production EKS clusters
- 6
- uptime
- 99.9%
- faster deploys
- ~20%
- lower database cost
- ~8%
- years in software
- 12+
- certifications
- 3
Recent reliability work
What running a healthcare platform's infrastructure has looked like this year.
Two fleet-wide Kubernetes upgrades, zero downtime
Led EKS 1.32 → 1.33 and 1.33 → 1.34 across all six production clusters with Helm, Helmfile and Codefresh — no customer-visible impact on either pass.
- EKS
- Helm
- Codefresh
Hardened a Lambda concurrency storm
A burst throttled a customer-facing webhook. Capped concurrency in Terraform, fixed the fail-fast path, and closed the detection gap with alerting on the unreserved pool.
- Lambda
- Terraform
- Alerting
Aurora Serverless v2 → RDS Provisioned
Migrated on CVE patch-latency grounds and cut monthly database cost by about 8%. Later found and killed three months of orphaned double billing.
- RDS
- PostgreSQL
- Cost
Field notes
Incident write-ups and migration notes from production — what broke, what I checked, what changed.
- The 504 That Wasn't My Server: A CloudFront VPC Origin OutageMy site went down with 504s and my first instinct was 'out of memory.' The server was fine — AWS's VPC-origin plumbing had silently stopped forwarding traffic. Here's how I proved it, worked around it, and reverted once AWS fixed it.
- Claude on Bedrock Wouldn't Bill in India — So I Shipped on NovaI had the feature designed, the prompt written, and the IAM key scoped. The one thing standing between me and a working AI feature was a payment method AWS wouldn't accept. Here's the blocker on the India region — and the model-agnostic escape hatch that let me ship the same day.
- What the First 50 Portfolios Taught Me About Building XIRR LedgerXIRR Ledger just crossed 50 real portfolios. Here's why I built it, what the early numbers actually mean for retail investors, and where the product goes next.
- Infrastructure Debt Shows Up on Your Credit CardWe migrated Aurora to RDS with zero downtime — then quietly paid for two databases for two months. The fix wasn't more knowledge. It was a forcing function.
Skills and tooling
- Cloud & infrastructure
- AWS (EKS, EC2, RDS)LambdaS3CloudFrontRoute 53KMSAPI Gateway
- Orchestration & IaC
- KubernetesDockerTerraformHelmCodefreshKyverno
- Observability
- PrometheusGrafanaThanosCloudWatchInfluxDBSLO / SLI design
- Languages
- PythonBashSQLPHPJavaScript
- Databases
- PostgreSQLMySQLAuroraRedisAWS DMS
- Security
- WizCheckovSnykHadolintAikido Safe-Chain
- Practices
- CI/CDIncident managementCost optimisationNetwork automationZero-downtime migrations
Products I've shipped
Live sites I designed, built and still operate — infrastructure included.
- CardKhataForward an Indian credit card statement by email and get every transaction categorised, EMIs rebuilt, fees itemised and fee-waiver progress tracked.
- XIRR LedgerServerless personal-finance tool computing true portfolio XIRR from broker ledgers, benchmarked against Nifty 50.
- QiyasDubai property price transparency from real Dubai Land Department transaction records — medians, price per sq ft, trends.
- Rudra AstroVedic astrology platform on self-designed AWS infrastructure with bulk email to 2M+ users via SQS and SES.
- Ganga PujaOnline puja booking from Gangotri — 500+ pujas by verified pandits with prasad delivery worldwide.
Running something that has to stay up?
I'm open to conversations about SRE and platform roles, and about infrastructure that needs a steadier hand. I reply within a day.