Remote work
that actually
pays.

Evaluate AI model outputs for accuracy, helpfulness, and safety. Compare responses and provide detailed feedback to improve model performance across multiple domains.

Label and classify images for computer vision training datasets. Work includes bounding boxes, segmentation, and categorical tagging for autonomous driving models.

Design, test, and refine prompts for large language models. Create evaluation criteria and test cases for model outputs. Work with cutting-edge AI systems.

Test AI systems for vulnerabilities and safety issues. Attempt to find edge cases, biases, and harmful outputs. Help make AI safer for everyone.

Curate and organize large-scale training datasets for Google's AI models. Ensure data quality, diversity, and proper labeling across text, image, and audio modalities.

Evaluate AI model outputs for ethical concerns, biases, and fairness. Provide detailed reports on problematic patterns and suggest improvements for more equitable AI systems.

Provide reinforcement learning from human feedback (RLHF) for advanced language models. Rank, rate, and compare model outputs to improve AI alignment and capability.

Review and validate medical AI outputs for accuracy and safety. Ensure AI-generated health information meets clinical standards and doesn't contain harmful advice.