Senior Software Engineer / AI Infrastructure & Data Quality Support — AI output review and LLM response evaluation
Performed LLM output review and quality assessment by checking accuracy, relevance, instruction-following, formatting, and safety for AI-assisted workflows. Identified hallucinations and evaluated structured results to flag quality, security, copyright, and data-leakage risks for workplace AI usage. Documented findings with precision using rubric-based judgment to support model and prompt improvement loops. • Rated responses and reviewed model-generated outputs for factuality and relevance. • Conducted safety review including hallucination detection, safety checks, and compliance-oriented validation. • Supported prompt-response quality improvement by investigating issues and root causes from logs. • Maintained technical documentation and rationales for evaluation outcomes and defects.