I am Just a Girl, trying to do some of the most difficult things, in the most interesting ways possible:)

I completed my Bachelor's degree in Mathematics in 2024 and spent 1 year post graduation doing National service (NYSC); working on National AI projects under the National Centre for AI and Robotics (See website here). Recently, with the outstanding success of current SOTA models and AI systems, I have spent time pondering on what I consider to be some of the most important problems facing AI and Humanity; how can we ensure that AI is safe, trustworthy, and works well for all of humanity (the global majority, global minority, and those in-between)? How can we ensure that we reduce present and future catastrophic risk (anticipated and unanticipated risks), that may arise from current and future advances in AI systems?

I am currently interested in Trustworthy and Safe AI (the technical bits, to be precise). I think of this in four different ways; the science of building models (model training and data integrity), the science of model alignment (AI Alignment and post training methods), the science of model evaluations (benchmarking), and the science of adoption (i.e, human-AI interaction, AI-Assisted decision making, AI agents, human centered AI, AI colonialism, governance, etc).

Incase it interests you, you should know that before now, I spent most of my formative years (during my undergrad and one-year post undergrad) working on theoretical machine learning; optimization for machine learning and the theoretical foundations of deep learning.

PS: I will be resuming my Mphil in Human-Inspired AI @ University of Cambridge, under the Centre for Human-Inspired Artificial Intelligence (CHIA), within the Institute for Technology and Humanity (ITH), this fall (October, 2026).