Research & Publications

Developing transformer-based language models for major African languages including Swahili, Amharic, Yoruba, and Zulu. This project aims to create high-quality language models that understand the nuances and cultural contexts of African languages.

Computer vision system for identifying diseases and optimizing yields in crops commonly grown across Africa. Using deep learning to analyze satellite imagery and field photos to provide actionable insights to farmers.

Fine-tuning models for sentiment analysis across multiple African languages, with a focus on social media text and news articles. Addressing the lack of sentiment analysis tools for African languages.

Natural language processing tools for extracting insights from medical records in African healthcare systems. Working with local hospitals to improve patient care through better data analysis.

Automated wildlife monitoring system using camera traps and computer vision to track animal populations across African national parks. Collaborating with conservation organizations.

Developing AI tools to personalize education for African students, including adaptive learning systems and automated essay grading in local languages.