System Design · OpenAI · Hard
Build a complete pipeline for detecting and moderating policy-violating videos on a large-scale platform. Your system must satisfy the following requirements: Recognize several policy categories, including violence, self-harm, hate, sexual content, and other harmful classes. Operate both when a video is uploaded and when scanning content that is already publicly available. Combine automated classification with human review. Reduce both missed violations (false negatives) and…
Checking your access…