Remote AI Data Ops Coordinator (Freelance)

Rex.zone; United States; On-site; USD 41,328 - 68,880; Full time

Job summary

Rex.zone in the United States is seeking a remote freelance generalist to support multiple workstreams that connect product delivery with AI/ML training workflows. The role blends execution, coordination, and quality control to help teams maintain training data quality and adhere to annotation guidelines.

You will coordinate with engineering, data operations, and vendors, organize training datasets, perform QA on labeling outputs, and contribute to model performance improvements while

Qualifications

• Mid-Senior level in operations or delivery roles supporting technical teams.
• Familiarity with AI/ML data workflows and quality frameworks.
• Strong written communication for remote stakeholders.
• Experience across NLP and CV annotation domains.
• Ability to reason about edge cases and guidelines.

Responsibilities

• Coordinate cross-functional work in a remote environment.
• Organize, triage, and validate training datasets and labeling outputs.
• Perform QA evaluation on labeled data and RLHF data.
• Enforce annotation guidelines and update guidelines based on edge cases.
• Support NLP and CV annotation tasks, including NER and segmentation review.
• Contribute to content safety labeling workflows and escalation.
• Track throughput, accuracy, rework, and propose process improvements.
• Maintain documentation, handoffs, and status updates for distributed teams.

Skills

Operations experience

Remote collaboration

Documentation skills

Edge-case reasoning

Multidomain adaptability

Job description

Rex.zone in the United States is seeking a remote freelance generalist to support multiple workstreams that connect product delivery with AI/ML training workflows. The role blends execution, coordination, and quality control to help teams maintain training data quality and adhere to annotation guidelines.

You will coordinate with engineering, data operations, and vendors, organize training datasets, perform QA on labeling outputs, and contribute to model performance improvements while