Muhammed Yusuf Dada

Muhammed’s MSc. research focused on applying model alignment techniques to study the biases, guardrail bypass, and privacy leaking tendencies of Large Language Models (LLMs). His goal was to uncover, report and analyze the risk factors in adopting and integrating LLMs as their popularity soars so that decision makers can make well-informed decisions on model choice and parameter configuration. Prior to beginning his Master’s, Muhammed worked full-time for 3 years across Software Engineering and Data Science roles applying NLP and LLM techniques (such as contextual embeddings) to industry use-cases.