video1.0<iframe src="https://www.loom.com/embed/c342ded2b38e48dca322eae76cecc09f" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/c342ded2b38e48dca322eae76cecc09f-a90f3ff475fb2e47.gif255.336Hi, I am Akshay, and I built a Crime Cause Card that prioritizes triage and safety instead of selling e-commerce products. I send prompts to Grok, with an added quotient to re evaluate the problem, and the UI uses green for normal and red for critical medical emergencies. It suppresses all commercial vehicles, and only shows comfort essentials with doctor level guidance. It supports English and Arabic, uses a three stage LLM pipeline with Grok and Llama, and validates outputs with Pydantic to prevent malformed JSON. It was proved with an automated eval suite of 12 cases. No viewer action was requested.