Industrial Physicist · AI/ML Engineer in Training
Building intelligent systems at the intersection of physics, climate, and machine learning
BSc Industrial Physics graduate from Covenant University with research in atmospheric signal propagation, machine learning for energy systems, and climate-driven wireless communication reliability. Based in Nigeria — working toward a Master's abroad in AI/ML Engineering.
background
I am a BSc Industrial Physics graduate (Electronics & IT Application) from Covenant University, Nigeria, where I graduated with a CGPA of 4.16/5.00. My academic formation spans electromagnetics, signal processing, atmospheric physics, and computational methods.
My research examines how climate-driven changes in atmospheric water vapour affect microwave radio wave communication in Abuja, Nigeria — applying ITU-R P.453-14 formulations to a decade of meteorological data and validating results against standard propagation models.
Beyond signal propagation, I develop machine learning systems for energy forecasting — including an ongoing Nigeria Grid Energy Demand Forecasting project using LSTM neural networks and Random Forest models trained on NERC grid data.
I was a selected participant in the Climate Change AI Virtual Summer School 2026, where I completed tutorials in statistical downscaling, optimal power flow with AI, tsunami alert prediction, and bioacoustic monitoring — all notebooks publicly available on GitHub.
technical toolkit
selected work
An ongoing ML project developing electricity demand forecasting models for Nigeria's national grid. Built Linear Regression and Random Forest baselines, then developed an LSTM neural network in PyTorch/Keras for time-series demand prediction. Data sourced from NERC and Our World in Data. Streamlit dashboard deployment in progress.
Four complete tutorial notebooks from the Climate Change AI Virtual Summer School 2026, covering: statistical downscaling of climate projections with DeepESD, AI for optimal power flow, tsunami alert prediction from earthquake observations, and agile bioacoustic species monitoring. All publicly available on GitHub.
professional history
research output
This study applies the ITU-R P.453-14 formulation to a decade of meteorological observations for Abuja, Nigeria (2015–2024) to compute monthly and seasonal profiles of water vapour pressure, saturation vapour pressure, and enhancement factors. A Multiple Linear Regression model validated against the ITU-R P.453-14 propagation standard achieved an R² of 95.94%, identifying atmospheric water vapour as the primary predictor of signal attenuation. Seasonal peaks in April–May present the highest risk to frequencies above 10 GHz, with implications for 5G and satellite link reliability in West African tropical climates.
get in touch
Open to research collaborations, graduate supervision inquiries, and scholarship opportunities. Feel free to reach out directly.