QA Engineer

Crossing Hurdles; South Africa; On-site; ZAR 300,000 - 420,000; Full time

Job summary

Crossing Hurdles is seeking a quality assurance specialist focused on AI validation and data labeling in South Africa. You will design tests for AI outputs and document results. You will annotate datasets, evaluate model predictions, and collaborate with cross-functional teams to improve data quality and model performance.

RLHF experiences and robust testing practices are valued. The role emphasizes meticulous documentation, clear communication, and continuous improvement of AI systems for

Qualifications

• Hands-on expertise in software quality assurance with a strong understanding of test planning and defect reporting.
• Experience with data labeling, annotation, RLHF, or model evaluation.
• English proficiency at B2 level or higher for collaboration and reporting.

Responsibilities

• Design, execute, and document test cases to validate AI outputs against requirements.
• Review, annotate, and label data sets to improve model training accuracy.
• Evaluate and report on model predictions, identifying inconsistencies and errors for improvement.
• Collaborate with cross-functional teams to share findings and clarify requirements.
• Contribute to AI-specific testing methodologies and best practices.
• Maintain detailed records of evaluations and test results with clear feedback.
• Provide actionable recommendations to improve data quality and model performance.

Skills

Test planning

Defect reporting

Data labeling

RLHF evaluation

Model evaluation

English communication

Job description

• Design, execute, and document test cases and scenarios to validate AI outputs against defined requirements.
• Review, annotate, and label data sets to enhance machine learning model training accuracy.
• Evaluate, analyze, and report on model predictions, identifying inconsistencies, errors, or patterns for continuous improvement.
• Collaborate with cross‑functional teams to communicate findings, raise questions, and clarify project requirements.
• Contribute to the development and refinement of testing methodologies specific to AI‑driven systems.
• Maintain detailed, organized records of evaluations, test results, and feedback through written documentation.
• Provide actionable recommendations to optimize data quality and model performance.

Requirements

• Hands‑on expertise in software quality assurance, with a strong understanding of test planning and defect reporting.
• Strong relevant experience with data labeling, annotation, reinforcement learning from human feedback (RLHF), or model evaluation.
• Proficiency in English at B2 level or above for effective collaboration and reporting.