Available for Work

Data Scientist · ML Engineer · Economist

AREMU GIDEON
MARVELOUS.

Fraud Detection · Credit Risk · Macroeconomic Forecasting

I turn raw data into decisions as an ML engineer with an economist's lens, blending domain expertise with statistical modeling in a way that sets me apart.

Published Researcher·International Grant Recipient·Production ML Systems

MacroSense

Actual vs Predicted Unemployment Rate (6 Months Ahead)

Live
ActualXGBoost ForecastEnsemble ForecastRecession
Actual0.0XGBoost Forecast0.0Ensemble Forecast0.0

The actual US unemployment rate against 6-month-ahead forecasts from an XGBoost model and an ensemble model. The shaded band marks the recession period highlighted in the chart.

01 / About Me

THE PERSON
BEHIND THE MODELS.

I'm a data scientist and economist who builds models and understands why every decision behind them matters, from the validation method to the cost of a wrong prediction.

My work spans Predictive Modelling, Forecasting, fraud detection, credit risk Analysis, and econometrics, and I take each model past the notebook into a live, deployed system.

I'm also a researcher with published papers and international research grant. I bring a researcher's statistical rigour and an engineer's follow-through.

3+Years Experience
8+Projects Completed
5+Clients Worked With

02 / Selected Work

SELECTED WORKS.

A curated collection of projects I've built, from fraud detection to macroeconomic forecasting, each taken past the notebook into something decision-ready.

Click View Project on any card for the full case study.

Fraud DetectionImbalanced Learning

Fraud Detection ML

A machine learning system that scores credit card transactions for fraud risk in real time.

Code
Credit RiskExplainable AI

CreditIQ

A machine learning approach to predicting loan default with explainable AI.

Code
ForecastingDeployed App

MacroSense

A machine learning system for forecasting US economic indicators using Federal Reserve data.

Code
EconometricsResearch

Nigeria Macro Modeling

Econometric vs machine learning approaches to modeling Nigerian macroeconomic indicators.

Code

03 / Tools & Technologies

TOOLS & TECHNOLOGIES.

A well-rounded stack I rely on to build explainable, production-grade machine learning and forecasting systems.

Machine Learning & Statistics

XGBoost logoXGBoost
Random Forest
Logistic Regression
LightGBM
SHAP Explainability
Isolation Forest
PyTorch logoPyTorch
TensorFlow / Keras logoTensorFlow / Keras
SMOTE
ARIMA
GARCH
ARDL
Bayesian Calibration
Causal Inference

ML Engineering & MLOps

Python (OOP, Pipelines, Testing) logoPython (OOP, Pipelines, Testing)
FastAPI logoFastAPI
Streamlit logoStreamlit
Docker logoDocker
MLflow logoMLflow
DVC
CI/CD (GitHub Actions) logoCI/CD (GitHub Actions)
SQL
MongoDB logoMongoDB
pandas logopandas
NumPy logoNumPy

Data & Tools

FRED / Kaggle / AlphaVantage APIs logoFRED / Kaggle / AlphaVantage APIs
ETL Pipeline Design
Feature Engineering
Git logoGit
AWS S3 logoAWS S3
GCP logoGCP
Power BI logoPower BI
Matplotlib / Seaborn

04 / Testimonials

WHAT CLIENTS SAY.

Kind words from people I've had the pleasure of working with.

Gideon turned a vague brief into a fraud detection solution that paid for itself within weeks. Clear updates at every step, and the final report was the most professional analytics work I have received from any freelancer.

HC

Hilda Corp

Freelance Client, UK

The first project went so well that I came straight back to him with a bigger one, and he still over-delivered. Communication was flawless, deadlines were met early, and the results were beyond what I paid for.

HC

Hilda Corp

Freelance Client, UK

Working with Gideon felt like having a senior data team on demand. He handled everything from messy data to a deployed model, explained every choice in plain English, and my stakeholders were genuinely impressed.

BR

Babatunde Rodiat

Freelance Client, US

05 / Contact

LET'S WORK TOGETHER.

Have a project in mind or a problem you'd like to explore? Send me a message and let's talk.