Jatin Kumar

Applied prompt engineering methods such as prompt rewriting, optimization, and reverse prompting to test and improve AI model accuracy and responsiveness. Evaluated multimodal AI outputs (text, images, audio, and video) against safety, quality, and platform compliance guidelines. Identified and categorized toxic, harmful, biased, or inaccurate AI outputs to strengthen safety guardrails and model alignment. Performed contextual QA audits on AI-generated transcriptions and media for grammatical correctness and factual alignment.

Managed concurrent live chat sessions to resolve customer inquiries using CRM tools with high accuracy and policy compliance. Leveraged internal knowledge bases to deliver timely, accurate, and professional customer resolutions.

Reviewed and moderated high-volume user-generated text and image content to enforce platform policies and community safety guidelines. Used critical judgment for sensitive, flagged, and high-risk material across different categories. Detected and removed policy-violating submissions efficiently to support a safer user experience.