Data annotation projects on Mindrift. Help build better AI, from wherever you are.
Mindrift Data Annotation is where contributors help train and improve the AI models behind everyday technology. This page is home to our data annotation projects — the most accessible side of Mindrift. These projects cover things like labeling, categorizing, transcribing, and reviewing data. They're designed to be approachable: no specialized background is needed to get started, just attention to detail and consistency. (For projects that call for specific expertise — coding, STEM, legal, and more — visit the main Mindrift page.) Contributors pick the projects that fit their schedule and take part remotely, from anywhere. Every contribution feeds directly into building AI that's more accurate, more useful, and more representative of the people who rely on it. Want to contribute to AI development on your own terms? Explore current projects and join the Mindrift community.
Remote · Markham, Ontario, Canada
Record short, casual voice clips in English and Chinese, naturally switching between the languages while responding to prompts or reacting to short videos. Complete several clips per recording session and ensure recordings are clear, with minimal background noise and only one speaker.
Remote · Calgary, Alberta, Canada
Record short, casual voice clips in English and Chinese, naturally switching between the languages in everyday conversation. Respond to prompts or react to short videos and submit several clips per recording session.
Remote · Toronto, Ontario, Canada
Record short voice clips responding to prompts or reacting to videos, speaking naturally in English and Chinese. Complete several clips per recording session and ensure the recordings are clear and free of background noise.
Remote · Canada
Record short, casual voice clips responding to prompts or reacting to videos, naturally switching between English and Chinese. Complete several clips per recording session and ensure recordings are clear, close to the microphone, and free of background noise.
Remote · Canada
Record short voice clips, typically 1–2 minutes, in response to prompts or short videos. Speak naturally while mixing English and Chinese, and complete several clips per recording session.
Remote · Canada
Review and label text, image, or video data based on specific project guidelines. Identify and flag factually incorrect, sensitive, or inappropriate AI-generated content.
Remote · Canada
Review and label text, image, or video data based on specific project guidelines. Identify and flag factually incorrect or inappropriate AI-generated content.
Remote · Burnaby, British Columbia, Canada
As an AI Trainer - Annotator, you will evaluate AI-generated responses and provide feedback to improve AI models. Your contributions will shape how AI systems understand and interact with real-world tasks.