• Development and maintenance of web and mobile frontend applications using React and React Native with Expo • Design and implementation of serverless backend services on AWS, using Lambda and DynamoDB • Integration between frontend and backend through scalable and high-performance APIs

• Led adversarial and benign prompt engineering initiatives to rigorously stress-test large language models • Performed fine-grained classification and risk assessment of user inputs across safety-critical content categories • Evaluated model outputs for alignment with responsible AI principles, policy compliance, and misuse prevention • Identified failure modes and edge cases in model behavior, contributing to iterative safety improvements • Applied content moderation frameworks to proactively mitigate harmful or disallowed AI responses • Generated structured evaluation data to support model alignment, robustness, and trustworthiness objectives • Partnered with internal safety guidelines to ensure consistent enforcement of ethical and regulatory standards