Movement & Physical Activity Expert Annotator

In this role, you'll evaluate AI-generated content for scientific accuracy, safety, and relevance while reviewing and annotating datasets related to physical activity, exercise, and movement patterns. Your work may involve wearable sensor data, activity logs, video, transcripts, and other multimodal data that support the development of next-generation AI models.

COMPENSATION

• Up to $50/hour for U.S.-based contractors

• 40 hours per week (8 hours/day)

• At least 4 hours of overlap with Pacific Time (PST)

• Fully remote

• 8-week contractor engagement

• Immediate start available

WHAT YOU'LL DO

• Annotate activity type, intensity, duration, and movement quality across diverse data sources

• Review wearable sensor data, activity logs, videos, transcripts, and related datasets

• Apply standardized movement and exercise taxonomies (e.g., MET values, exercise classifications, gait and posture labels)

• Evaluate AI-generated responses for scientific accuracy, safety, and clinical relevance

• Identify inconsistencies between self-reported activities and objective data signals

• Flag ambiguous, incomplete, or low-quality data for further review

• Collaborate with domain experts to refine annotation guidelines and improve edge-case handling

• Consistently meet quality, accuracy, and throughput expectations

WHAT WE'RE LOOKING FOR

We're seeking professionals with:

• 3+ years of experience in a relevant field

• A background in Exercise Science, Kinesiology, Physical Therapy, Sports Science, Human Performance, or a related discipline

• Familiarity with physical activity classification systems and wearable sensor data

• Experience interpreting human movement, biomechanics, or exercise programming

• Strong analytical skills and attention to detail

• Excellent written communication and documentation skills

Successful candidates will also demonstrate:

• The ability to apply detailed annotation guidelines consistently

• Comfort working with spreadsheets, labeling platforms, or annotation tools

• Strong critical thinking and pattern recognition skills

• The ability to work independently in a remote environment

PREFERRED QUALIFICATIONS

While not required, experience in any of the following is a plus:

• Exercise physiology or biomechanics research

• Wearable technologies (Fitbit, Garmin, Apple Watch, WHOOP, Polar, etc.)

• Motion capture, gait analysis, or movement assessment

• AI evaluation, human data annotation, or model benchmarking

• Clinical rehabilitation, sports performance, or athletic training

• Machine learning datasets involving health, movement, or sensor data

CONTRACT DETAILS

• Positions available: 15

• Employment type: Contractor assignment/freelancer 

• Duration: Short-term contract (8 weeks)

• Start Date: Immediate

• Commitment: 8 hours per day, with an overlap of 4 hours with PST

• Location: USA 

Please Note

• A MacBook is ideal for completing this work.

• Candidates should be comfortable working with spreadsheets, annotation tools, or labeling platforms as part of daily task delivery.

Verita engagements are remote and asynchronous, and Verita says they are paid weekly.