Generative AI-Powered Automated Workload Evaluation with Python, CloudWatch, and Amazon Q
summary
A Python script using Boto3 + AWS Organizations and AWS CLI scans multi-account, multi-region environments with STS role assumption and multithreaded listing, normalizes resources into workload units using workload unit ratios 1 / 0.1 / 0.02 (EC2 / Fargate / Lambda), logs CSV/text, and publishes custom metrics to Amazon CloudWatch dashboards. Amazon Q generates summaries, pinpoints issues (missing CloudWatch agent, LightSail SSH open), produces and runs a Python remediation script (installs pip deps), and the recorded demo shows the automation actually works end to end.
AI-written summary · about
chapters (23)
- 0:00Introduction and presenters
- 1:47Why this use case matters
- 3:57Agenda and AWS services
- 5:02AWS Organizations and CloudWatch
- 6:08Workload types explained
- 7:11Script requirements and permissions
- 8:55Vendor comparison and in-house solution
- 10:19Multi-account and multi-region support
- 11:21Script features and customization
- 13:29Workload ratio calculations
- 14:55Logging and output formats
- 15:36Prompt mode and interactive options
- 16:18CloudWatch dashboard visualization
- 17:25Introducing Amazon Q Developer
- 18:09Amazon Q Developer capabilities
- 20:41Live demo with Amazon Q
- 21:01Running the script via Amazon Q
- 22:30Generating Terraform and JSON
- 23:12Deep dive into workloads
- 24:18Fixing security issues with prompts
- 25:00Script review and execution
- 26:30Leveraging LLM for automation
- 27:14Conclusion and resources
related talks
21mAWS On Air ft. Amazon Inspector
34mEmpowering AWS DevOps with Python and Machine Learning
43mAI-Driven Quality Engineering Practices
35mThe Good, the Bad and the Ugly — one year of integrating AWS security services
12mLeveraging AWS and Kubernetes for Scalable AI/ML Workloads in the Cloud
14mBattle-tested tactics to accelerate data science and ML R&D with AWS
15mVMware Cloud on AWS: Automation and integration, the real power of API
44mBest practices for analytics and generative AI on AWS(ANT302)