Browse categories
Explore
Fiverr Pro
English
$
USD
Abdullah Ahmed
@abdullah_o_zaki
Turning Complex Data into Actionable Business Insights ML , BI Expert
Egypt
Arabic, English
About me
Results-driven Data Analyst & ML Engineer with a proven track record of building production-grade analytics solutions. I specialize in transforming raw, messy data into clear, profitable business decisions.
My Impact:
- Built an AI Real Estate platform with 99.5% prediction accuracy analyzing 17,000+ listings
- Delivered BI solutions analyzing 74.9M SAR in revenue across multiple branches
- Resolved 3,400+ complex data quality issues via automated ETL pipelines
- Expert in Python Machine Learning (XGBoost, SHAP) and enterprise Business Intelligence (Power BI, DAX, Star Schema)
Skills
Abdullah Ahmed
Offline •
See my services
Machine Learning
I will build end to end machine learning and predictive analytics
Data Dashboards
I will design professional interactive power bi dashboards and bi solutions
Portfolio
Work experience
Self-Employed / Freelancer
Self-employed • 2 mos
Independent Data Science Consultant
Jun 2026 - Jul 2026 • 1 mo
Built and delivered an end-to-end AI analytics platform for the UAE real estate market as an independent consultant. Key achievements: - Analyzed 17,042+ property listings across 8 emirates via an automated Python ETL pipeline into PostgreSQL (Supabase). - Trained an XGBoost regression model reaching 99.5% prediction accuracy (R2 = 0.9950, MAPE 3.0%). - Applied SHAP explainability to prove location drives 84% of property value. - Delivered an interactive Streamlit dashboard with 6 modules, deployed live with real-time database connection. - Identified 4 buyer personas using K-Means clustering + PCA for targeted marketing strategies.
Independent BI Consultant
Apr 2026 - May 2026 • 1 mo
Built and delivered an end-to-end AI analytics platform for the UAE real estate market as an independent consultant. Key achievements: - Analyzed 17,042+ property listings across 8 emirates via an automated Python ETL pipeline into PostgreSQL (Supabase). - Trained an XGBoost regression model reaching 99.5% prediction accuracy (R2 = 0.9950, MAPE 3.0%). - Applied SHAP explainability to prove location drives 84% of property value. - Delivered an interactive Streamlit dashboard with 6 modules, deployed live with real-time database connection. - Identified 4 buyer personas using K-Means clustering + PCA for targeted marketing strategies.
