Post a Job Free
105 viewsjobseeker
Jane M. — Mid-Level AI Trainer & LLM Evaluator from South Africa

Jane M.

Mid-Level AI Trainer & LLM Evaluator

South Africa 6+ years
Open to offersNew to Platform
Languages
English
Video Introduction
No video introduction yet
The candidate has not added a video.
Contact information and social networks are private. Connect to unlock.
Hidden

About

Jane M. is a seasoned Shona Language Specialist, AI Trainer, and LLM Evaluation Specialist with over three years of hands-on experience in the artificial intelligence and consumer goods sectors. She specializes in prompt formulation, rubric design, and Shona linguistic annotation, delivering high-signal training data and evaluation frameworks that enhance AI model accuracy and cultural alignment. Her work includes crafting task-specific prompts for shopping decision-making, designing and applying weight-based rubrics, and detecting error patterns in AI-generated content. Jane has contributed to AI model development at organizations such as Prolific Academic, TELUS AI, Pareto AI, and TechInnovate AI, focusing on multilingual, multimodal quality assurance, RLHF ranking tasks, and structured review of AI outputs across text, image, and video data.

Experience

  • AI Prompt Designer & Rubric Evaluator

    Freelance | Independent Contractor · 2024 — Present
    Develop precise prompts tailored for consumer goods shopping and purchasing decisions, focusing on product comparison and feature prioritisation. Create evaluation rubrics with specific success and failure criteria, assessing dimensions like relevance, accuracy, completeness, and neutrality. Evaluate AI-generated outputs using these rubrics, pinpointing systematic failures and offering structured feedback for enhanced purchasing-decision outputs. Refine prompt designs based on rubric performance data to improve response quality and neutrality.

  • AI Evaluator – Multimodal QA

    TELUS AI | Freelance · 2022 — 2024
    Assess text-to-image and video outputs for semantic consistency, cultural bias, and content quality. Utilize structured rubrics to evaluate large-scale multilingual batches, identifying patterns of bias and instances of quality failure.

  • AI Trainer & Data Annotation Specialist

    Pareto AI | TechInnovate AI · 2019 — 2023
    Oversee data preparation for NLP and computer vision projects, employing tools such as CVAT, SageMaker, and Prodigy. Ensure model accuracy by consistently providing high-quality annotated datasets. Create annotation guidelines for tasks including text classification, named entity recognition, and image labeling.

  • Shona Language Specialist – AI Trainer & LLM Evaluator

    Prolific Academic | Freelance · 2022 — Present
    Conduct reviews and annotations of Shona AI outputs, focusing on grammar, fluency, semantics, and cultural suitability. Evaluate LLM outputs using established rubrics, identifying hallucinations and factual inaccuracies. Document recurring errors and provide corrective feedback to development teams. Support RLHF fine-tuning processes by ranking AI-generated responses based on quality preference.