Nana Sam Yeboah

I am Nana Sam Yeboah, a research engineer specializing in Natural Language Processing (NLP) and deep learning applications in healthcare and education. My work is driven by the goal of leveraging machine learning to advance clinical AI and enhance access to education. My research interests include multimodal medical generative AI models, and developing robust, reliable, interpretable AI systems that extend medical expertise across resource constrained settings.

Currently, I am a Machine learning Engineer at Minohealth AI Lab, where I build multimodal LLMs designed to assist doctors and healthcare practitioners in their work, such as through the Foundation Model for Radiology. My research also explores drug therapeutics and utilizes LLMs to delve into the field of genomics.

In addition to my professional work, I am actively involved in Brilla AI, an open-source initiative aimed at providing equitable access to education in Ghana. Through this project, we are developing an AI system to compete in and win the National Science and Maths Quiz (NSMQ). As the lead of the Text-to-Speech and Question Extraction teams, I have spearheaded the development of a high-accuracy speech synthesis model and created efficient pipelines to process real-time audio from contests, extracting relevant questions.

My contributions have been recognized at various conferences and workshops, where I have presented my work at notable events such as the Deep Learning Indaba and AfricAIED. Additionally, I have developed and shared valuable tools and models that are used in the community, reflecting my commitment to open-source work and collaborative research.

In my leisure time, I enjoy exploring legal thrillers, playing UNO with friends and playing football. I am committed to advancing the field of AI and am eager to collaborate with like-minded researchers and mentors in the area of clinical AI and genomics to contribute to impactful projects.