Computer Vision for Industrial Inspection (Training Project)
→
Summary
Completed hands-on training and applied industrial computer vision workflows for image classification and defect detection.
About
Highly accomplished Data Scientist and Machine Learning Engineer specializing in developing and deploying end-to-end predictive systems across healthcare, retail, and industrial AI domains. Proven track record of delivering measurable impact, including improving forecast accuracy by 25% and reducing reporting overhead by 40% through scalable ML pipelines and automation. Adept in Python, ensemble modeling (XGBoost, LightGBM), explainable AI (SHAP), SQL-based data extraction, feature engineering, and robust model deployment.
Work
Thandel Integrated Concept Ltd
|Data Scientist
Remote
→
Summary
Led the development and deployment of predictive machine learning models to enhance sales forecasting and optimize operational efficiency for data-driven inventory decisions.
Highlights
Designed and deployed predictive machine learning models, significantly improving sales forecasting accuracy by up to 25% and enabling data-driven inventory decisions.
Optimized data preparation by querying and transforming structured datasets using advanced PostgreSQL techniques (JOINs, aggregations, window functions) to create training-ready datasets for ML pipelines.
Architected comprehensive end-to-end ML pipelines, streamlining data extraction, preprocessing, feature engineering, model training, evaluation, and serialization processes.
Automated data processing workflows in Python, reducing manual reporting effort by 40% and enhancing operational efficiency.
Implemented robust ensemble models, including Random Forest and XGBoost, which reduced RMSE by 15% compared to baseline regression models.
Applied SHAP for advanced model interpretability, communicating critical feature-level insights to business stakeholders to support informed, data-driven decision-making.
Salvation Ministries
|Electrical Engineer
Port Harcourt, Rivers State, Nigeria
→
Summary
Managed and executed electrical system installations and upgrades for large-scale facilities, significantly enhancing operational capacity and ensuring safety compliance.
Highlights
Led critical electrical system installations and infrastructure upgrades for large-scale facilities, directly increasing operational capacity by 20%.
Reduced overall project costs by 30% through strategic optimization of procurement planning and implementation of lean execution strategies.
Developed and implemented structured safety and compliance protocols, resulting in a flawless record of zero reportable safety incidents across all projects.
Education
WorldQuant University
New Orleans, LA, United States of America
→
Certificate (In Progress)
Applied Data Science
Osiri University
Lincoln, NE, United States of America
→
MS
Data Science & Information Systems
Rivers State Polytechnic
Bori, Rivers State, Nigeria
→
Higher National Diploma (HND)
Electrical/Electronic Engineering
Rivers State Polytechnic
Bori, Rivers State, Nigeria
→
National Diploma (ND)
Electrical/Electronic Engineering
Languages
English
Certificates
Computer Vision for Industrial Inspection
Issued By
NVIDIA
Deep Learning Certificate
Issued By
Simplilearn
Kubernetes for Developers (LFD259)
Issued By
Linux Foundation
Kubernetes for Cloud-Native Applications (LFS250)
Issued By
Linux Foundation
Skills
Programming
Python (Pandas, NumPy, Scikit-learn, XGBoost, LightGBM), SQL, PostgreSQL.
Machine Learning
Regression, Classification, Ensemble Methods, Cross-Validation, Hyperparameter Tuning, Model Evaluation (RMSE, MAE, R2, ROC-AUC).
Data Engineering
PostgreSQL (JOINs, aggregations, filtering, window functions), ColumnTransformer, OneHotEncoder, StandardScaler, SelectKBest, SMOTE.
Model Explainability
SHAP.
Deployment
Streamlit, Model Serialization (Joblib, Pickle).
Tools
Git, GitHub, Jupyter Notebook, Linux, Kubernetes (training).
Projects
Diabetes Risk Prediction and Deployment
→
Summary
Developed and deployed an end-to-end classification model to predict diabetes risk from clinical and lifestyle features, leveraging LightGBM, Scikit-learn, and Streamlit.
Income Classification and Model Interpretability
→
Summary
Led the development of an income classification model for a U.S. Census dataset, incorporating XGBoost, SMOTE, and SHAP for enhanced accuracy and interpretability.
Supermarket Sales Forecasting
→
Summary
Developed a high-accuracy regression model using XGBoost and ColumnTransformer for transaction-level supermarket sales forecasting.