Data Science Projects Portfolio

A structured monorepo for portfolio-grade data science work: reusable modeling code, notebooks, reproducible demos, and production-oriented references.

Start here (first 2 minutes)

1. Clone and bootstrap:

   git clone https://github.com/diogoribeiro7/ds-projects-portfolio.git
   cd ds-projects-portfolio
   python -m venv .venv
   .venv\Scripts\Activate.ps1  # Windows
   # or: source .venv/bin/activate  # macOS/Linux
   pip install -r requirements.txt

   Optional contributor stack:

   pip install -r requirements-dev.txt   # linting, type-checking, pre-commit, dev tooling

2. Run the portfolio demo:

   python examples/run_demo.py

3. Validate the repository check (optional but recommended):

   make check

4. Open a featured notebook to evaluate portfolio quality quickly:

   • notebooks/healthcare_analysis.ipynb
   • notebooks/finance_credit_risk_stress_testing.ipynb
   • examples/eda_end_to_end.ipynb

What this portfolio demonstrates

This repository is centered on practical projects where technical quality is paired with clear storytelling:

• Modeling reliability: healthcare explainability, credit-risk modeling, forecasting, and fraud detection.
• Deployment-readiness: FastAPI workflows, MLOps drift management, and notebook-to-service transitions.
• AI engineering: RAG, retrieval/eval service patterns, and controlled guardrails.
• Decision quality: calibration, subgroup fairness, benchmarking, and governance-oriented reporting.

Featured portfolio set

Use this short list first when reviewing:

• notebooks/healthcare_analysis.ipynb (healthcare)
• notebooks/life_science_medical_imaging_triage.ipynb (medical imaging)
• notebooks/insurance_data_science.ipynb (insurance governance)
• notebooks/finance_credit_risk_stress_testing.ipynb (finance stress testing)
• notebooks/genai_rag_pipeline.ipynb (GenAI production pipeline)
• notebooks/energy_load_probabilistic_forecasting.ipynb (time-series forecasting)
• notebooks/mlops_mlflow_drift_lifecycle.ipynb (MLOps lifecycle)

Project index

Use this map when you want to jump directly to a project area:

• churn_prediction
• customer segmentation
• ab_testing
• statistical_methods
• streamlit_apps
• dashboard_enhanced
• archive legacy MLOps notebooks
• deep_learning
• nlp
• performance_optimization

The complete notebook catalog (including archive/reference notebooks) remains in notebooks/README.md.

End-to-end analytics examples

• examples/eda_end_to_end.ipynb for a complete exploratory analysis walkthrough.

Repository layout

.
├── src/                  # Core reusable packages
├── docs/                 # Portfolio documentation and API reference
├── notebooks/            # Portfolio notebooks (featured + archive)
├── projects/             # Project writeups and reference implementations
├── examples/             # Runnable portfolio demos
├── tests/                # Automated test suite
├── scripts/              # Repo tooling
├── deployment/           # Portfolio-facing deployment entry points
└── archive/              # Historical artifacts and legacy references

Repository scope contract is documented in docs/PORTFOLIO_SCOPE.md.

Read next

• Portfolio visitors: docs/index.md and docs/README_ENHANCED.md
• Contributors: docs/internal.md, docs/contributor/development.md
• If you spot an issue: open it on GitHub

This project is meant to stay portfolio-first: fast signal for reviewers, concise entry paths, and clear separation between maintained portfolio content and archive backlog.