Understand how our policies, our detection systems, and our human enforcement fit together today, articulate a compelling picture of where image and text safety needs to be, and make the case for what has to change—in policy, in systems, or in process—to get there. Define how we measure this for every feature and every policy, and you own the goals. Turn today's one-off, feature-by-feature safety work into a shared system that any team at YouTube can plug into. Support fewer and better detection models instead of many overlapping ones. Make deliberate trade-offs about where to spend and where to simplify.
Job description provided by Google.
View the original posting
