Overview
We are currently inviting applications for an internship position focused on building an agentic system for cultural tailoring of multi-modal content for the African Health Stories project. Africa Health Stories is a collaborative project between Microsoft Research and Stellenbosch University, the University of Pretoria and Swansea University to create a story generation tool which produces interactive visual, oral, and text stories enabling patients to better understand how to apply health advice to their local circumstances.
This role presents a unique opportunity to contribute to the development of Machine Learning (ML) and Image Generation technologies that are culturally and linguistically grounded in the African context.
Qualifications
Required:
- PhD in Computer Science, Machine Learning, Statistics, Engineering, Mathematics, Physics, or related field or equivalent experience.
- Enrolled in, or recently completed, doctoral studies in Computer Science, Machine Learning, or related field.
- Knowledge of state-of-the-art and emerging Language Models, including their application in complex systems.
- Familiarity with Large Language Models such as the OpenAI GPT models, Phi3.5 etc., model fine-tuning techniques (LoRa, QLoRa), and quantization, prompting techniques (Chain of Thought, ReACT etc.), and Retrieval Augment Generation
- Coding and engineering skills to design and run experiments and help to drive research into product.
Preferred:
- Strong understanding of foundational LLM architectures and optimization techniques
- Experience with multimodal models and AI agentic systems.
- A passion for bridging research and real-world impact, with a desire to see innovations translated into practical solutions.
- Publication record in relevant conferences, such as ACL, EMNLP, NeurIPS, ICLR, ICML, etc.
