About the Employer Insure Organisation is South Africa’s fastest growing short‑term insurance company, committed to creating a nation of great drivers through its innovative Shared‑value Insurance model. It offers comprehensive products in the motor and home insurance sectors and runs the internationally recognised Vitality Drive incentive programme. Over 1 000 people work across the organisation, focusing on unique solutions and customer centricity. Overview Working as an intermediate Data Scientist, you will develop and integrate natural language processing (NLP) and large language models (LLMs), with a particular focus on image models. These capabilities will enable autonomous decision‑making, improve operational efficiency, and deliver personalised customer experiences. Key Responsibilities Collaborate with content specialists, data engineers and system architects to deploy solutions and deliver projects from inception through to business adoption, focusing on Image AI. Deploy robust and scalable LLMs integrated into AI frameworks, enhancing NLP capabilities. Incorporate Retrieval‑Augmented Generation (RAG), vector databases and semantic searches to improve precision and relevance in data extraction from large datasets. Design and implement automation workflows that improve speed and accuracy of processes, reducing manual intervention and operational costs. Develop performance metrics to evaluate AI model efficiency and accuracy; monitor and maintain systems to ensure they remain effective and adaptable. Ensure compliance with laws and ethical guidelines for AI models and data handling; prepare documentation and reports for regulatory bodies. Prototype ML systems and AI concepts, especially using NLP and LLMs, and evaluate model performance. Monitor advances in AI and LLM technologies and review academic literature and industry releases to keep strategies aligned with best practice. Required Experience 2–4 years in a data science environment. Demonstrated ability to implement ML workflows at scale, particularly using LLMs. Experience handling, analysing and extracting insights from large, complex datasets, especially unstructured text. Previous experience tuning open source and proprietary large language models. Additional experience with image AI models (e.g. damage detection) is advantageous. Desired Experience 5 to 10 years. Required Education Matric (essential). Bachelor’s degree in Computer Science, Mathematics, Statistics, Data Science, Actuarial Science, Operations Research, Industrial Engineering, Applied Mathematics or a comparable quantitative field. Honours or Master’s degree in a relevant field is advantageous. Desired Qualification Level Degree. Required Technical Skills SQL and database working. Python for data science and machine learning. Knowledge of TensorFlow, PyTorch, NLP and LLM libraries. Familiarity with Azure services. Familiarity with Databricks. Desired Skills SQL and working with databases. Python for data science and machine learning. Familiarity with Databricks. Familiarity with Azure services. #J-18808-Ljbffr