Six hands-on sessions that take you from a first deploy to a working AI-powered application. You build in your own AWS account with Kiro as your AI pair, you learn to check what it suggests, and our Lab Checker verifies the result against your real resources.
Every session is practical. You leave with something running that you built, and a way to check that you built it correctly.
You work in a real AWS account and deploy real resources. The teaching happens while you build, not before it.
Enter your bucket, instance or role and the checker inspects your account with read-only calls, stamping each step pass or fail.
Every failed check points you back to the exact step to revisit, so you fix it yourself instead of waiting for a demonstrator.
Cert-aligned and job-ready. Each session builds on the last, and finishes with a lab the checker can verify.
Set up Kiro, then ship something real on day one: buckets, static hosting, public access policy, and how a request actually reaches your page.
You leave with: a live public website and a working AI-assisted development setup.
Instances, key pairs and security groups, and what to do when it does not answer on port 80. You practise asking the AI the right question, then proving the answer.
You leave with: a running web server and the habit of verifying what the AI tells you.
Trust policies, permissions policies, and why wildcard and admin access is the habit that breaks accounts. The AI is good at drafting policy and confidently wrong about scope.
You leave with: a least-privilege role and the IAM vocabulary an interviewer expects.
Lambda, DynamoDB and API Gateway wired into an endpoint that returns real data. The AI writes the scaffolding fast; you read every line before it goes anywhere.
You leave with: a live API you can explain line by line.
Add a model call to the API you already built, and put the guardrails on before the bill arrives: a spending limit, a token cap and a clear idea of what each request costs.
You leave with: an endpoint that returns model output, with cost controls in place.
Put it together: a frontend, the API, your model call, and a CloudWatch alarm that actually tells someone when it breaks.
You leave with: a deployed full-stack AI application, a working alarm, and a two-minute demo you can show an employer.
It only inspects. It never creates, changes or deletes anything in your account, and it runs with a role limited to read-only actions. You can open the check files and see exactly what it does.
Cloud bills come from things you forgot to delete. After teardown, run the checker again: a red FAIL on step 1 is your proof of a tidy account, and that is the point.
Free, hands-on and run in Sydney. Create an account to use the Lab Checker and watch the recordings afterwards.
Join on Meetup