Remote Data Scientist: AI Model Evaluation & Quality
Posted Updated
Description
TELUS Digital is seeking a data scientist to safeguard data quality and push AI insights. You’ll design evaluation pipelines to measure model accuracy, bias, and safety across large datasets, and wrangle data for LLM training. You’ll detect failure patterns, visualize trends, and deliver actionable guidance for ongoing model improvements.
You will also automate data-quality checks on human-annotated data to uphold TELUS Digital standards, while working remotely as part of a global AI community.
Architect and execute evaluation frameworks to measure model accuracy, helpfulness, and safety across datasets. Perform advanced data-wrangling to transform raw text into optimized training sets for Large Language Models (LLMs). Detect and visualize trends in model failures, providing actionable insights for AI training. Develop automated scripts to monitor the consistency of human-annotated data, ensuring quality standards. Bachelor's degree in Data Science, Mathematics, Statistics, Physics, or a related technical field. English Proficiency: minimum B1 (Intermediate) level English to communicate data trends. Native-level fluency in at least one language. Basic proficiency in data manipulation with Python, SQL, or Excel.Skills
Sign in to read and join the discussion.