Multilingual Image & Text AI Quality Expert

About the role

Join the TELUS Digital AI Community and make a direct impact on the machine learning models of some of the world’s largest brands. As a Multilingual Image & Text AI Evaluator, you will participate in a specialized two-stage visual evaluation project for generative AI models. You will evaluate AI-generated images for text accuracy, identify visual distortions, and perform precise localized corrections to ensure AI outputs meet high linguistic and visual standards.

Responsibilities

• Text Correctness Scoring: Evaluate and score text embedded within AI-generated images for spelling, grammar, vocabulary, and contextual accuracy.
• Localized Bounding-Box Corrections: Perform precise spatial bounding-box annotations to capture, flag, and correct spelling errors, typographical issues, and visual rendering distortions.
• Guideline & Quality Adherence: Consistently learn, apply, and maintain complex visual and linguistic annotation guidelines across all assigned tasks.
• Feedback & Escalation: Identify subtle visual/textual anomalies, flag ambiguous data, and ensure high annotation trustworthiness across project workflows.

Requirements

• Thai
• French (France)
• Vietnamese
• Malay
• Japanese
• Korean
• Indonesian
• Arabic (Gulf)
• Spanish (Mexico)
• Portuguese (Brazil)
• Tagalog