Sai tarun Kadari

I am a Senior AI Quality Analyst focused on evaluating large language model outputs, detecting hallucinations, and ensuring bias and safety compliance. I bring a strong technical foundation in Python, SQL, and NLP fundamentals to structured evaluation workflows.

I have reviewed over 1,000 AI-generated responses across technical and general domains, consistently achieving 98%+ accuracy adherence to guidelines and improving evaluation alignment scores. I design and refine prompts to increase model response reliability and clarity.

My hands-on experience includes building an ANN-based predictive model with full dataset preprocessing, validation, and performance metric evaluation. I implement fact-check workflows, structured validation checklists, and defect logging processes that increase QA reliability in fast-paced environments.

I work effectively with remote collaboration tools and apply QA methodologies—test case creation, dataset sampling with SQL, and validation scripting in Python—to deliver measurable improvements in model behavior and evaluation consistency.