About The Role
Support AI/ML development by producing and evaluating high-quality training data for real engineering and technology workflows. Work includes data labeling for NLP and computer vision, prompt evaluation for large language model behavior, RLHF-style preference judgments, and QA evaluation to ensure annotation guidelines compliance.
What You Will Do
- Execute labeling and evaluation tasks across text, image, and multimodal datasets
- Perform named entity recognition, content classification, and content safety labeling
- Review model outputs during large language model evaluation and document edge cases
- Run QA sampling, discrepancy analysis, and track error types to improve training data quality
- Collaborate with remote teams to refine rubrics, guidelines, and acceptance criteria
Workflow and Tools
- Work in structured queues with rubrics, golden sets, and calibration sessions
- Use web-based annotation platforms, spreadsheets, and basic scripting for checks
- Maintain versioned guidelines and inter-annotator agreement targets
Required Skills
- Strong STEM fundamentals, careful reading, and consistent decision-making
- Basic Python and SQL; attention to detail and clear written communicationComfort interpreting structured rubrics and ensuring guidelines compliance
Employment Details
- Workplace: Remote
- Employment Type: Full-time
- Compensation: $30–$50 per hour (base salary)
How To Apply
Create or update your Rex.zone profile, highlight STEM coursework or projects, and emphasize relevant skills (data labeling, QA evaluation, RLHF, NLP, computer vision annotation, content safety labeling). Apply through Rex.zone and be prepared for short guideline checks and quality reviews.
#J-18808-Ljbffr