24/7 Python CLI on AWS with claude ai

I need my existing Python-based CLI and the accompanying Claude code deployed on AWS so the script can run non-stop, recover automatically from any failure, and alert me the moment something goes wrong. At the moment I do not have an AWS account in place, so part of the job involves guiding me through account creation and choosing the right services. Once live, I expect the solution to execute the script continuously, watch for errors around the clock, and restart or scale as required without manual intervention.

Key deliverables
• Set up a fresh AWS account (or walk me through it) and configure the necessary IAM roles and security settings
• Provision the infrastructure—EC2, Lambda, or another service you recommend—to host and keep the Python CLI running 24/7
• Implement automated error monitoring and alerting via CloudWatch, SNS, or equivalent so I’m notified immediately on failure
• Provide an easy way to redeploy updated code (Git, CodePipeline, or Docker image)
• Document the architecture and hand over all credentials, run-books, and setup steps so I can maintain or extend the system later

Acceptance criteria
• Script runs continuously for 48 hours without manual restarts
• Any forced error triggers an alert within one minute
• All infrastructure is reproducible via IaC (Terraform or AWS CloudFormation) and captured in version control

If you have solid AWS DevOps experience, especially with long-running Python processes and automated monitoring, I’d love to hear how you would tackle this.