📍 Lahore, Pakistan 🇵🇰
"I build ML systems that solve real problems — from healthcare risk prediction to e-commerce intelligence."
profile = {
"name" : "Muhammad Usman",
"role" : "Data Scientist | ML Engineer | Analytics Engineer",
"education" : "BS Data Science (In Progress)",
"location" : "Lahore, Pakistan 🇵🇰",
"focus" : [
"Machine Learning & Predictive Modeling",
"Data Analytics & Business Intelligence",
"AI Engineering & Model Deployment",
],
"currently_building" : "Intelligent data pipelines + deployed ML models",
"learning" : ["Deep Learning", "MLOps", "LLM Applications"],
"goal" : "Build AI systems that create measurable business impact",
"open_to" : ["Internships", "Research Collaborations", "Part-time DS roles"],
}I'm a Data Science student with hands-on experience building end-to-end ML pipelines — from raw data wrangling to deployed Streamlit dashboards. My projects span predictive modeling, business analytics, and healthcare AI. I focus on writing clean, reproducible code and shipping projects that actually work.
Healthcare ML system predicting diabetes risk using clinical measurements. Built with a professional multi-notebook architecture and reusable src/ modules for production-ready code.
- Tech Stack: Python, Scikit-Learn, Jupyter
- Signal: Healthcare AI domain + clean code architecture
Full analytics pipeline turning raw e-commerce data into business intelligence using Python, Pandas, Plotly, and Power BI. Delivers clear insights on sales patterns and KPIs.
- Tech Stack: Python, Pandas, Plotly, Power BI
- Signal: Business analytics + BI tools = industry-ready
End-to-end ML pipeline with MySQL database backend and an interactive Streamlit web dashboard. Regression models predict property prices with real-time user input.
- Tech Stack: Python, MySQL, Streamlit, Scikit-Learn
- Signal: Database integration + deployment = rare for students
Business intelligence ML project using Logistic Regression and Random Forest to identify at-risk customers — enabling proactive retention strategies for real business impact.
- Tech Stack: Python, Scikit-Learn, Pandas
- Signal: Classic industry ML use case with business framing
📍 NOW — May 2026
├── ✅ Python (Advanced) — Variables, OOP, Functions, Modules
├── ✅ Pandas & NumPy — Data manipulation & numerical computing
├── ✅ ML Algorithms — Regression, Classification, Clustering
├── ✅ Data Visualization — Plotly, Matplotlib, Seaborn, Tableau
├── ✅ Business Intelligence — Power BI, Dashboards
├── ✅ Database Integration — MySQL, SQLite
├── ✅ Model Deployment — Streamlit web apps
├── 🔄 End-to-End ML Pipelines — Feature engineering, cross-validation
└── 🔄 Project Architecture — src modules, reproducible notebooks
🎯 NEXT — 6 Months
├── 🧠 Deep Learning (TensorFlow / PyTorch)
├── 🤖 Natural Language Processing & Transformers
├── ☁️ Cloud Deployment (AWS SageMaker / GCP Vertex AI)
├── 🔧 MLOps — Model monitoring, CI/CD, versioning
└── 📦 LLM Applications & AI Engineering
🚀 FUTURE GOALS
├── 💼 ML Engineer / Data Scientist at a tech company
├── 📦 Build & deploy production AI SaaS products
└── 🌍 Contribute to open-source AI/DS projects
| Opportunity | Status |
|---|---|
| 🎓 Data Science / ML Internships | ✅ Open |
| 🔬 Research Collaborations | ✅ Open |
| 💼 Part-time DS / Analytics Roles | ✅ Open |
| 👥 Open Source Projects | ✅ Open |