AI Technical Trainer & Data Contributor
Evaluated AI-generated code outputs for accuracy and safety within the Aether project framework. Ranked and scored model responses based on adherence to complex reasoning and formatting guidelines. Identified and corrected logical errors in Python, Java, and C++ snippets to improve response quality. • Reviewed code correctness and safety constraints • Performed evidence-based QA using detailed justifications • Assessed adherence to reasoning/format requirements • Flagged issues and recommended corrections