Movement & Physical Activity Annotator

Braintrust; United States; Hybrid; USD 41,000 - 55,000; Full time

Job summary

Braintrust is seeking experienced exercise, movement, or physical-activity professionals to support AI training and evaluation by labeling activity, intensity, and movement data from sensors, videos, and transcripts.

Role requires 3+ years in related fields, familiarity with wearable data, and meticulous labeling practices. The position spans 8 weeks with 8-hour days and overlaps Pacific Time for coordination.

Qualifications

• 3+ years in exercise science, kinesiology, physical therapy, sports science, fitness coaching, rehabilitation, or related field.
• Familiarity with physical-activity classification and wearable or sensor data.
• Strong attention to detail and following structured labeling guidelines.
• Comfort using spreadsheets, annotation tools, or labeling platforms.
• Ability to commit for 8 weeks, work 8 hours per day, and overlap at least 4 hours with Pacific Time.

Responsibilities

• Annotate activity type, intensity, duration, and movement quality from sensor streams, logs, video, or transcripts.
• Apply taxonomies such as MET values, exercise categories, gait labels, and posture labels.
• Compare self-reported activity with objective signals and flag incomplete or low-quality data.
• Review AI-generated exercise and movement content for scientific accuracy, safety, and relevance.
• Help refine annotation guidance and edge-case handling while meeting quality and consistency expectations.

Skills

Exercise science

Kinesiology

Data labeling

Education

CPT/EP-C/PT/DPT credentials

Tools

Annotation platforms

Job description

Our client is seeking experienced exercise, movement, or physical-activity professionals to support an AI training and evaluation initiative. You will review and label activity, exercise, and movement-pattern data so health-focused AI systems can be evaluated for accuracy, safety, and relevance.

What You'll Do

• Annotate activity type, intensity, duration, and movement quality from sensor streams, logs, video, or transcripts.
• Apply established taxonomies such as MET values, exercise categories, gait labels, and posture labels.
• Compare self-reported activity with objective signals and flag incomplete, inconsistent, or low-quality data.
• Review AI-generated exercise and movement content for scientific accuracy, safety, and relevance.
• Help refine annotation guidance and edge-case handling while meeting quality and consistency expectations.

Requirements

• At least 3 years of relevant experience in exercise science, kinesiology, physical therapy, sports science, fitness coaching, rehabilitation, sports performance, or a related field.
• Familiarity with physical-activity classification and wearable or sensor data.
• Strong attention to detail and comfort following structured labeling guidelines.
• Comfort using spreadsheets, annotation tools, or labeling platforms.
• Ability to commit for 8 weeks, work 8 hours per day, and overlap at least 4 hours with Pacific Time.

Preferred Qualifications

• CPT, EP-C, PT/DPT, kinesiology, sports-science, or related credentials or education.
• Experience with accelerometer, heart-rate, GPS, gait, or posture data.
• Prior health/fitness annotation or basic AI/ML data-pipeline familiarity.

Compensation

$35 USD per hour.