Not the typical data scientist origin story.
I didn't start with a laptop and a Kaggle account. I started with a blackboard, 40 students, and a burning question: Why are we still teaching like it's 1995?
That question followed me for 15 years across classrooms in Lagos and Ogun State — teaching Mathematics, Further Mathematics, Physics, and Chemistry. And somewhere in the middle of all that chalk dust, I found data science.
In late 2024, I joined the Federal Government's 3MTT programme.
Something clicked.
I realised that the same clarity I used to break down quadratic equations for JSS3 students was the same clarity I needed to think through a machine learning problem. Teaching had already trained me. I just needed to aim it differently.
So I did.
I build ML-powered solutions — mostly around education, business risk, and social impact. I also run HMG Concepts, a virtual educational institution I founded in 2015, with three subsidiaries: HMG Academy, HMG Technologies, and HMG Media.
My work sits at the intersection of data science and human development. I don't separate the two. Never have.
Core stack: Python · Pandas · NumPy · Scikit-learn · Matplotlib · Seaborn SQL · Excel · Power BI · Tableau · Git/GitHub
Currently deepening: Deep Learning & NLP (DSN × Google.org DeepTech_Ready, Cohort 3) Applied Data Science (WorldQuant University) ML Engineering (Kodecamp Cohort 6)
| Project | What It Does |
|---|---|
| Insurance Claim Prediction | Predicts fraudulent insurance claims using Random Forest + SMOTE + SHAP |
| SwiftChain Delivery Prediction | Multi-class classification of delivery delay risk for logistics |
| Bank Customer Churn | Identifies customers likely to exit using ensemble models |
| Yakub Promotion Prediction | Predicts staff promotion eligibility from HR data |
| Income Level Prediction | Binary classification of income brackets from census-style data |
| Fake News Detector | NLP-powered classifier distinguishing real from fabricated news |
| Burnout Prediction | Predicts employee burnout risk from workplace indicators |
| Student At-Risk Predictor | Flags students likely to fail based on academic + behavioural data |
| Student Performance Tracker | Tracks and visualises student academic performance over time |
| Student Study Plan Generator | AI-powered personalised study plan builder for students |
| CBT Question Bank | Structured question bank management system for CBT exams |
| CBT Pro | Full CBT system (teacher + student portals) used with real students |
I don't build by accident. Every project above started with a real problem I observed — in a classroom, in a dataset, in a conversation.
I leverage AI as a thinking tool, not a shortcut. The architecture decisions, the problem framing, the feature engineering — that's deliberate work. Always.
"Learning Deliberately. Teaching Authentically."
I believe the best data practitioners are also good teachers — people who can explain what they built and why it works. Fifteen years in the classroom gave me that. Data science gave me the tools to go further.
- 🌐 Portfolio: cssadewale.pages.dev
- 🏫 HMG Concepts: hmgconcepts.business.site
- 📺 YouTube: youtube.com/@hmgconcepts
- 📧 Email: buildingmyictcareer@gmail.com
Based in Lagos / Ogun State, Nigeria. Open to data science roles, collaborations, and EdTech partnerships.