Multilingual Specialist — LLM/VLM Evaluation & SFT Expert

We are currently seeking language experts with native or fluent proficiency in: Thai French (France) Vietnamese Malay Japanese Korean Indonesian Arabic (Gulf) Spanish (Mexico) Portuguese (Brazil) Tagalog About the Role: At TELUS Digital, we help the world’s leading organizations build safer, smarter, and more reliable AI systems. As a GenAI Multilingual Specialist — LLM/VLM Evaluation & SFT Expert, you will support the development and evaluation of next-generation Large Language Models (LLMs) and Vision-Language Models (VLMs). Your role is to act as the “human reasoning layer” behind AI systems by evaluating multilingual AI-generated content for factual accuracy, logical consistency, instruction adherence, and cultural relevance. You will work across multilingual Supervised Fine-Tuning (SFT), rubric-based scoring, fact verification, and multimodal (image + text) evaluation workflows. Responsibilities Evaluate AI Outputs Review AI-generated responses for quality, reasoning accuracy, factual correctness, and compliance with project guidelines. Create and evaluate multilingual Supervised Fine-Tuning (SFT) data to improve AI model performance across multiple languages and domains. Perform Multimodal (VLM) Assessment Analyze relationships between images and text outputs to evaluate contextual accuracy and multimodal reasoning quality. Apply Rubric-Based Scoring Assess outputs using structured scoring frameworks and standardized evaluation rubrics. Conduct Research & Fact Verification Verify claims using authoritative sources and apply citation standards to distinguish factual information from unsupported assertions. Escalate Edge Cases Identify ambiguous outputs, policy concerns, and reasoning inconsistencies that require escalation or additional review. Maintain Operational Accuracy Follow complex project SOPs and quality standards consistently across high-volume evaluation workflows. Qualification path What You Bring (Mandatory Qualifications): Language Expertise: Native or fluent-level proficiency in at least one required language with strong reading and writing ability. Analytical Thinking: Ability to evaluate content objectively and identify inconsistencies, logical gaps, or quality issues. Attention to Detail: Strong procedural discipline and the ability to follow detailed project guidelines accurately. Communication & Reliability: Strong communication skills with the ability to work independently in remote environments. Research & Verification Skills: Comfortable conducting fact-checking and validating information using credible and authoritative sources. Technical Familiarity: Previous experience in freelance work, translation, writing, annotation, AI data projects, localization, or content evaluation is a plus. AI/ML Experience: Prior experience with AI/ML annotation, LLM evaluation, or SFT projects is strongly preferred. Specialized Knowledge Areas Preferred Candidates with academic or professional backgrounds in the following areas are highly valued: Computer Science Biology Law Finance Physics Mathematics Engineering

The role involves evaluating AI-generated content for quality and accuracy, as well as creating multilingual Supervised Fine-Tuning data to enhance AI model performance. Additionally, the specialist will analyze multimodal outputs and conduct fact verification.

Candidates must have native or fluent proficiency in one of the specified languages and possess strong analytical and communication skills. Experience in AI/ML projects and familiarity with technical processes are preferred.

Relevées dans la description du poste. Confirmez les exigences importantes ci-dessus.

• Language Expertise
• Analytical Thinking
• Attention to Detail
• Communication
• Research Skills
• Verification Skills
• Technical Familiarity
• AI/ML Experience

Expérience minimale

0+ ans