//aws · terraform · claude code
INFRABOT
Claude-powered CLI that reads your AWS account state and generates production-ready Terraform modules. Cuts infrastructure review from 3 days to under an hour.
View project →Cloud DevOps engineer. AI builder. Solutions architect. 7+ years designing AWS infrastructure, automating the work that slows teams down, and shipping tools that actually get used.
//field notes | 06.27.2026 | edition 01
//aws · terraform · claude code
Claude-powered CLI that reads your AWS account state and generates production-ready Terraform modules. Cuts infrastructure review from 3 days to under an hour.
View project →//github actions · claude api · devsecops
AI gate inside GitHub Actions. Catches IAM over-permissions, security misconfigs, and config drift before anything reaches production. No manual review needed.
View project →//claude code · aws · cost optimisation
AI-assisted audit framework for cloud teams. Scans your AWS account, surfaces real findings, and produces a prioritised fix plan with estimated cost savings.
View project →7+ years in Cloud DevOps. I know what it looks like when the scope keeps growing — platform, cloud, DevOps, MLOps — more complexity, same team, no slack in the system.
I started building AI tools to handle the work that slows engineers down. Claude Code and the Anthropic API let me ship automations in days that would have taken weeks to build and months to maintain.
Specific problems. Real solutions.
AWS
EC2, Lambda, ECS, RDS, IAM, Cost Explorer. 7+ years hands-on.
Claude API
Tool use, agents, MCP servers, Claude Code automation.
DevOps
Terraform, GitHub Actions, Docker, Kubernetes, CI/CD.
Python
Automation scripts, API integrations, data processing.
Cloud DevOps specialist. AI builder. Solutions architect. I design the infrastructure, automate the manual work, and ship tools your team actually uses. If there is a bottleneck in your stack — I remove it.
engineer.ali91@live.comYes. Every engagement starts with an audit of what you already have. I build on top of your existing infrastructure — ECS clusters, Lambda functions, RDS instances, existing pipelines. Nothing gets ripped out unless it needs to be.
Your team should be building product, not writing runbooks and manually reviewing Terraform PRs. I build the agents that handle the repetitive layer — freeing your engineers for the work that actually moves the needle.
The first working automation ships within one week. I identify the highest-friction task in your pipeline, build the agent, integrate it, and hand it over — with documentation. Scope grows from there if you want it to.
Every tool I build ships with a rollback path, alerting, and documentation. I do not deliver black boxes. If it breaks, you know how to disable it in under a minute — and I am available to fix it. That is part of the deal.
//still evaluating?
LET'S TALK
YOUR STACK.
30 minutes. No pitch deck. Just your infrastructure and what is slowing it down.
Start the conversation →