MAHMOUD ISLAM DATA ANALYST & ML.
Data In, Decisions Out — Data Analyst & Machine Learning Engineer transforming raw data into actionable business insights. Building end-to-end analytics pipelines, calibrated ML models, and executive BI dashboards with Python, SQL, Power BI, and Scikit-learn across 100K+ records.
I am a Data Analyst and Machine Learning Engineer dedicated to building end-to-end analytics solutions that turn raw, messy data into sound business decisions. Pursuing my Bachelor's degree in Computer and Information Sciences at Minia University (2024–2028) with a Very Good cumulative grade, I bridge statistical thinking with scalable data engineering.
My analytics methodology spans 100K+ enterprise records: from relational database modeling (PostgreSQL, SQL Server) and automated SQL ETL pipelines to advanced feature engineering, calibrated machine learning (XGBoost, Scikit-learn), and executive business intelligence dashboards in Power BI and Tableau. I also deploy interactive prediction systems using Streamlit and Plotly Dash.
Awarded 3rd Place & Certificate of Excellence (Top 5) in the AI Hackathon Minya and certified in The Power of Statistics. My expertise is sharpened by rigorous training programs including The University of Tokyo (GCI World Data Science), DEPI (9-Month Specialist Track), NTI (98% Machine Learning, 94.5% Advanced Analytics), an internship at Elevvo Pathways, and 1,000+ algorithmic challenges solved across Codeforces and LeetCode.
COLLECT
CLEAN & EDA
FEATURE ENG
MODELING
EVALUATE
BI DASHBOARD
Problems Solved
Codeforces • LeetCode
AI Hackathon
3rd Place • Minya Award
Records Analyzed
Telecom • E-Commerce
Academic Grade
B.Sc. Minia University
ACCURACY
Leakage-safe feature pipelines and calibrated classification models.
SCALE
Optimized SQL queries, relational schemas, and automated workflows.
IMPACT
Data storytelling translating raw statistical noise into executive decisions.
My Tech Universe
Machine Learning & Model Evaluation
Developing end-to-end predictive systems with Python, Scikit-learn, and XGBoost. Specializing in leakage-safe feature engineering (180+ features), hyperparameter tuning, model calibration, and model interpretability with SHAP.
Data Analytics & Business Intelligence
Designing interactive executive dashboards in Power BI and Tableau using advanced DAX formulas. Performing deep exploratory data analysis (EDA), customer segmentation strategy (RFM & K-Means), and automated data cleansing.
Generative AI & Core Computing
Applied foundations in LLM Prompt Engineering and RAG architectures via ITI / NVIDIA DLI. Backed by relational database modeling in SQL and competitive programming discipline (1,000+ solved) in C++.
What I Do
Business Intelligence & Dashboards
Transforming complex, raw datasets into clean, executive-ready dashboards in Power BI and Tableau. Delivering interactive KPI cards, customer segmentation strategies, and clear data storytelling for non-technical stakeholders.
Predictive Modeling & Machine Learning
Designing, benchmarking, and optimizing high-precision classification and regression systems using Scikit-learn and XGBoost. Handling real-world imbalanced data, cross-validation, SHAP interpretability, and deployment via Streamlit.
Database Engineering & ETL Pipelines
Architecting relational star schemas and writing optimized SQL queries for high-performance data extraction. Automating data aggregation to eliminate manual reporting bottlenecks and maintain data governance.
Education & Training
Bachelor's degree in Computer and Information Sciences
Rigorous academic curriculum establishing theoretical computer science fundamentals and practical engineering practices. Emphasizing algorithmic problem solving, relational database management, probability & statistics for machine learning, data structures, and object-oriented architectures.
Advanced Data Analytics Summer Training
120-hour intensive program in applied analytics. Conducted Python exploratory data analysis (EDA) to cleanse, transform, and draw insights from raw records. Built Advanced Excel pivot tables and summary dashboards, designed interactive Power BI dashboards with DAX to monitor key business metrics, and evaluated machine learning classifiers (Logistic Regression, SVM, KNN, XGBoost) for customer churn forecasting.
GCI World: Data Science Theory & Implementation Program
Prestigious program covering theoretical data science rigor and end-to-end implementation. Successfully passed final assessment. Expanded the capstone into a full production-ready telecom customer intelligence product featuring 184 engineered features, a calibrated XGBoost model (ROC-AUC 0.69), SHAP explainability, K-Means customer segmentation, a Streamlit prediction tool, and a 5-page Plotly Dash BI dashboard.
Data Analysis Intern
Completed 4 real-world data analysis projects during an intensive practical internship. Ingested, cleaned, and explored diverse datasets, formulated hypothesis-driven analytical inquiries, derived actionable business recommendations, built clear visual reports, and fully documented reproducible workflows on GitHub.
Data Analyst Specialist Trainee
In-depth 9-month professional specialization covering the full data lifecycle with Python and SQL. Architected relational schemas, engineered automated SQL pipelines, and built executive BI dashboards in Power BI, Tableau, and Advanced Excel. Studied Data Governance and Prompt Engineering, delivered freelance client deliverables (database design, ML model creation, financial dashboards), and completed a comprehensive capstone platform.
Generative AI Training Program
Practical training focused on Python foundations for artificial intelligence, foundational Generative AI principles, LLM prompting strategies, and Retrieval-Augmented Generation (RAG) concepts through hands-on NVIDIA Deep Learning Institute curriculum materials.
Machine Learning Summer Program
120+ hours of hands-on machine learning training covering data preprocessing, leakage-safe feature transformations, supervised algorithms (trees, ensembles, linear models), hyperparameter optimization, model evaluation metrics (ROC-AUC, Precision/Recall, F1), and web deployment via Streamlit.
3rd Place & Certificate of Excellence (Top 5)
Awarded 3rd Place and a Certificate of Excellence (Top 5) in recognition of delivering an innovative artificial intelligence solution. Evaluated by expert judges on problem definition, data preparation, algorithmic architecture, model evaluation, and impactful presentation.
The Power of Statistics
Formal certification verifying mastery of statistical methods applied to data analytics and machine learning: probability theory, discrete and continuous distributions, hypothesis testing, confidence intervals, regression analysis, variance decomposition, and statistical inference.
Selected Work
Telecom Churn & Customer Intelligence Suite
Calibrated XGBoost • 184 Features • BI Suite • 2026End-to-end churn intelligence platform built on 100K customer records (originating from The University of Tokyo GCI World capstone). Engineered 184 features, trained and calibrated an XGBoost model achieving ROC-AUC of 0.69, and quantified feature importance via SHAP values. Performed K-Means customer segmentation, deploying both an interactive Streamlit simulation tool and a 5-page Plotly Dash BI dashboard.
E-Commerce & Customer Intelligence Platform
RFM Segmentation • Delay Prediction • Power BI • 2026Full-cycle intelligence platform analyzing 100K+ Brazilian marketplace orders (Olist). Segmented customer behavior via K-Means ($K=4$) RFM analysis, identifying "Champions" (29.8% of user base driving 57.5% of total revenue). Trained delivery-delay classifiers (ROC-AUC 0.774), dynamic Power BI drill-through dashboards, and a Streamlit scenario forecasting application.
Neuro Digital Analytics Platform
DEPI Graduation Project • Behavioral Analytics • 2026End-to-end analytics platform investigating the impact of short-form media habits on student wellbeing and academic focus. Engineered relational star schema SQL ETL, an interactive 5-page Power BI executive suite, and trained a Random Forest behavioral classifier (93.8% accuracy, 0.892 macro F1) deployed via Streamlit & 13-page Plotly Dash web tool.
Algorithmic Problem Solving Hub
Codeforces • LeetCode • Data Structures in C++ & PythonCentral repository of 1,000+ competitive programming solutions across Codeforces, LeetCode, and VJudge. Demonstrates mastery in Graph Theory, Dynamic Programming, Greedy paradigms, and complex data structures engineered with optimal space-time complexity in high-performance C++ and Python.
Initiate Contact
Let's build intelligent data solutions.
Open for data analytics, machine learning opportunities, collaborations, and freelance projects. Send a message or reach out through direct communication channels.