AI Cloud Systems Evaluator | DataAnnotation (2022–2023)
Created AI training and evaluation tasks for cloud engineering models focused on Terraform HCL, AWS CDK, CloudFormation, IAM policy design, and CI/CD troubleshooting. Designed realistic prompts with ambiguity and partial context to mirror engineering backlog conditions and expected validation outcomes. Built grading rubrics and reference solutions to assess correctness, completeness, maintainability, and security alignment of model-generated infrastructure code. • Prompt creation and scenario design with operational constraints • Reference (golden) solution authoring for infrastructure code • Output review of model-generated code and rubric-based scoring • Documentation of recurring failure patterns, edge cases, and consistency improvements