01
Loan Default ML Pipeline
Turning 148,670 loan records into a calibrated lending decision, and throwing away the first model because it was perfect.
2026Case study →
Engineer · Builder · Ecosystem builder
Building products, people and communities around problems worth solving.
I am an AI/ML engineer, an electrical & electronics engineering student and a co-founder. I turn hard problems into working systems, help other people build theirs, and start the communities that keep both going.
(01) — About the work
Build the thing. Then build what lets other people build.
What I do
01
Systems
Production machine learning and the backends around it. At GrandGale Technologies I lead AI/ML across two live platforms: demand forecasting, routing, NLP matching and fifty-plus FastAPI endpoints. On my own time, a calibrated credit-risk engine and a tax assistant that refuses to guess.
02
Hardware & ventures
Technology for problems that live outside a browser. AquaVolt pairs solar power with water purification for Northwestern Nigeria and went from 500+ applicants to a national top 10 and a stand at GITEX Nigeria. Sahara Deliveries is a last-mile logistics pilot for the Kano–Kaduna–Sokoto corridor.
03
Ecosystems
The infrastructure that lets other people build. I led Hult Prize at my university as a nine-stage pipeline rather than a pitch night, founded Innovate Club UDUS, and helped put 1,500 students in one marketplace at a campus trade fair.
The record
Numbers I can stand behind.
- 4.97 / 5.00
First Class
Electrical & Electronics Engineering, UDUS.
- 2nd
in Nigeria
Second-best IJMB A-Level result nationally, 2023.
- 148,670
loan records
Behind a calibrated credit-risk model serving a live API.
- 500+ → Top 10
AquaVolt
From 500+ applicants to national iHatch top 10.
- 1,500+
students brought together
One campus trade fair through the UDUS Entrepreneurship Club.
- Top 25%
of 2,000+ universities
Where Hult Prize UDUS ranked worldwide under my lead.
Organisations
Where the work has taken me.
GrandGale Technologies
AI/ML Engineer
GITEX Nigeria 2025
Exhibitor
Hult Prize
Innovation Lead, UDUS
RAIN Nigeria
Campus Lead
NITDA
Student Ambassador
Innovate Club UDUS
Founder & Lead
Baze University
Applied AI, Distinction
Usmanu Danfodiyo University
EEE, First Class
(02) — Selected work
Four projects, four ways of building.
A production machine-learning system, a hardware venture, an AI product and a campus ecosystem. Each one written as a story rather than a README.
solar-powered purification02
AquaVolt
A solar-powered water purification system for communities where the grid is unreliable and the water is not safe.
2025Case study →
03
TaxWiz
A Nigerian tax assistant that does the arithmetic deterministically and lets the language model do only what it is good at: reading.
2026Case study →
04
Hult Prize UDUS 2026
Not a pitch competition. A nine-stage entrepreneurship pipeline built from scratch at a university in Northwestern Nigeria.
2026Case study →
(03) — Workshop
On the bench right now.
01Experiment2026
Optical neural network, in NumPy↗
A photonic neural network simulated from the physics up: Mach-Zehnder interferometers as the trainable unit, transfer matrices for the layers, MNIST as the test.
02Experiment2026
Quantum randomness validation↗
A benchmark that puts pseudo-random, simulated-quantum and real quantum entropy sources through the same battery of tests and scores them.
03Analysis2026
Nigeria crude-oil export intelligence↗
Fifteen years of export data, a log-log elasticity model, GARCH volatility and a trade-finance gap estimate, written up as the kind of report a development bank would commission.
(04) — Notes
Writing that came out of the work.
01Machine learningAugust 2026
Why the first model was thrown away↗
Three model families scored a perfect ROC-AUC of 1.0000. The investigation into a missingness pattern that matched the target on all 148,670 rows, and what the honest number turned out to be.
02FairnessAugust 2026
What it costs to exclude gender and age from a credit model↗
Four and a half basis points of ROC-AUC. A fair-lending analysis that measures outcomes across groups the model never sees, using the adverse impact ratio as a review trigger rather than a legal test.
03DocumentationAugust 2026
Model card: mortgage probability of default↗
What is modelled, what is assumed, and what the model must not be used for. The most important distinction on the card is between the PD that was estimated and the LGD that was not.
(05) — Contact
Have a problem worth building around?
I am open to conversations about AI systems, engineering, product, energy, innovation programmes and the kind of ambitious idea that needs someone who will actually build it.