Find the toil
Identify a repetitive DevOps or cloud task, like triaging alerts, reviewing infrastructure changes, or writing runbooks.
Cloud and DevOps engineer
building on AWS
Open to DevOps and Cloud roles
Based in Karnataka, India
I build automated, secure AWS infrastructure, and I stay on until it runs without drama.
My storySELECTED WORK / 2024—26
Systems built
to run.
Infrastructure, delivery and reliability — from the network edge to the release pipeline.
01 KUBERNETES / PLATFORM ENGINEERINGProgressive Delivery PlatformA self-healing release platform built around GitOps, canary analysis and automated rollback. AWS · EKS · ARGO CD · TERRAFORM 02 SERVERLESS / IDENTITY AUTOMATIONEmployee Onboarding & IdentityAn AWS onboarding system connecting employee identity, workflow orchestration and document handling. AWS · COGNITO · STEP FUNCTIONS 03 CLOUD NETWORKING / WEB ARCHITECTUREThree-Tier AWS ApplicationA segmented AWS network separating the public entry point, application runtime and data layer. AWS · VPC · EC2 · RDS 04 RELEASE ENGINEERING / SERVERLESSZero-Downtime API DeploymentsBlue-green environments and controlled canary traffic for safer API releases. API GATEWAY · LAMBDA · CLOUDWATCH
BUILT WITH INTENT. OPERATED WITH CARE. MORE ON GITHUB ↗
CAPABILITIES / TOOLKIT
What I work
with.
The tools I use to build, automate and operate cloud infrastructure — and the technologies I'm actively exploring.
01
Cloud infrastructure
Amazon Web Services EC2VPCIAMS3RDSDynamoDBLambdaAPI GatewayECS & FargateCloudFrontRoute 53Step FunctionsCognitoSNSSESSecrets ManagerCloudWatch 02
Containers & orchestration
Docker 
Kubernetes Amazon EKS KarpenterGateway APIIngress 03
Infrastructure as code
Terraform CloudFormation 
Ansible 04
CI/CD & GitOps
Git 
GitHub Actions 
Jenkins 
Argo CD Argo Rollouts 05
Observability
Prometheus 
Grafana LokiAlertmanager 
OpenTelemetry 
Jaeger 06
Security
IAM least privilegeSecurity GroupsSecrets managementPolicy as code
07
Linux & networking
Linux 
Bash DNSTCP/IPSubnetting 
Nginx 08
Reliability engineering
Incident responseRunbooksSLOsAlertingRollback strategy
09
AI & agent systems Exploring
Claude API 
OpenAI API Tool callingMCP 
LangGraph RAG basicsAmazon BedrockGuardrails & approval gatesAgent evaluationToken & cost control 10
Programming & interfaces
Python YAMLJSONSQLREST APIs PROCESS / HUMAN-LED AUTOMATION
AI does the toil.
I build the agents
that do it safely.
Give the agent one job
A narrow role, clear tools, and read-only access by default.
Connect real tools
The agent works through AWS, Kubernetes, GitHub and Slack, not just through chat.
Keep a human in charge
Anything destructive needs approval, and every action is logged.
Measure and improve
I test it against real incidents and track accuracy, cost and time saved.
AGENT ROADMAP / CLOUD OPERATIONS
Meet the
agents.
A team of focused agents for the repetitive work around cloud infrastructure. Each starts with a narrow job, scoped tools and a human approval boundary.
01 Incident TriageReads alerts and logs, then posts a likely cause and next steps to Slack. 02 Terraform ReviewerReviews a plan before it’s applied and flags risky changes. 03 Release GuardianWatches a canary release and recommends continue or roll back. 04 Cost WatcherFinds idle or oversized AWS resources and suggests savings. 05 Security AuditorFlags over-broad IAM permissions and open security groups. 06 Runbook WriterTurns an incident timeline into a clear, actionable runbook.
A LITTLE ABOUT ME
Hi, I’m Amit.
I’m a Cloud & DevOps Engineer building the systems behind reliable products.
My path into cloud engineering started with materials engineering at NITK, took me through manufacturing and teaching, and led me to building on AWS. I enjoy making complex systems understandable, automating the repetitive work, and designing infrastructure that people can depend on.
More about me Have a cloud project
in mind?
Start a project
Tell me what you’re building and when you need it. I’ll get back to you as soon as I can.
amitpm.work@gmail.com
Prefer LinkedIn? Let’s connect.