Muhammad Aqeel

• Built an end-to-end customer churn prediction pipeline using Python, XGBoost, and Pandas, identifying the top
20% of at-risk customers for targeted retention.
• Achieved 81% model accuracy (F1-score) on the churn classification task through iterative feature engineering and
hyperparameter tuning.
• Deployed production-ready ML models as REST APIs using Flask and Docker, integrated into AWS pipelines for
near real-time scoring.
• Processed and encoded 200, 000+ raw data records, handling missing values and categorical feature encoding to im
prove downstream model performance.
• Documented experiments and workflows in Jupyter Notebooks and established a reproducible model versioning pro
cess.

• Collected, cleaned, encoded, and preprocessed structured and unstructured datasets for machine learning and busi
ness applications, improving data quality and pipeline throughput.
• Performed exploratory data analysis (EDA) and feature engineering to surface actionable business insights and re
duce model training time.
• Designed, trained, evaluated, and optimized machine learning and deep learning models for classification, regression,
forecasting, and anomaly detection with production-ready pipelines.
• Built end-to-end ML pipelines using Python and industry-standard libraries; containerized services and supported
deployment to AWS environments.
• Wrote reusable, well-documented code and unit tests following software engineering best practices to ensure repro
ducibility and maintainability.
• Collaborated with product and engineering teams to translate business requirements into data-driven, AI-powered
solutions and monitored model drift in production.

• Cleaned, encoded, and analyzed datasets for academic research projects, supporting multiple publications and im
proving dataset readiness for modeling.
• Assisted in implementing machine learning models and dashboards to communicate research findings to non-technical
stakeholders.
• Automated repeatable preprocessing steps and documented data dictionaries to accelerate future research workflows.

• Applied data mining, analytics, and visualization techniques to extract actionable business insights for government
training and evaluation projects.
• Collaborated in cross-functional teams to translate analytical findings into policy recommendations and operational
improvements.

• Analyzed and visualized government training datasets using Python and Power BI to evaluate program effectiveness
and outcomes.
• Built and optimized SQL queries to extract and preprocess large datasets for analysis, reducing query runtime and
manual processing.
• Delivered data-driven reports to management to support policy and program improvement decisions.