Design, build, and scale content safety systems including classifiers, vector databases, and multimodal models to protect business-critical products and GenAI experiences. Develop and maintain production-grade distributed systems and content processing pipelines optimized for high throughput and reliability across server-side and on-device environments. Train, evaluate, and productionize models, incorporating feedback loops and automation to continuously improve model quality and performance.
Implement agentic workflows and advanced heuristics for deep threat understanding, enabling the proactive detection of complex abuse patterns. Drive agile engineering efforts to identify and mitigate novel abuse patterns, ensuring Google’s products remain engaged and safe in a shifting threat landscape.
Job description provided by Google.
View the original posting
