Skip to content

Denis0242/CareFlow

Repository files navigation

🚀 Featured Project: CareFlow — Causal Inference & Product Analytics for Decision-Making

🎯Live App: Live App: https://jn4vvly3pf2qrnfkrgka2t.streamlit.app

GitHub: https://github.com/Denis0242/CareFlow


📌 Overview

CareFlow is an end-to-end analytics project built with Python and Streamlit to demonstrate how causal inference, experimentation thinking, machine learning, and interpretable analytics can support better product and business decisions.

This project is positioned around real-world Product Data Analyst and Data Analyst workflows: analyzing outcomes, identifying drivers, validating relationships in data, tracking performance, and translating findings into actionable recommendations.


🎯 Business Problem

Teams often make decisions based only on correlation, which can lead to weak conclusions, poor prioritization, and low-confidence recommendations.

CareFlow addresses that gap by combining structured analytics with causal thinking to help answer higher-value questions such as:

  • What factors are most strongly associated with outcomes?
  • Which relationships are meaningful enough to investigate further?
  • How can data be used to support smarter, evidence-based decisions?
  • How can analytics improve prioritization for product, growth, and strategy teams?

🔍 Product Analytics Focus

  • User and outcome behavior analysis
  • KPI tracking and performance monitoring
  • Trend and pattern identification
  • Decision-driven analytics
  • Experimentation and causal thinking
  • Data-driven decision support

📊 Key Features

  • Causal inference and relationship analysis
  • Machine learning model comparison
  • SHAP-based model interpretability
  • Interactive Streamlit application
  • Notebook-driven exploration and validation
  • Reusable scripts for analytics workflows
  • Automated testing for reliability

🧠 Analytical Approach

  • Explored structured data to understand patterns, distributions, and outcome relationships
  • Applied causal inference concepts to move beyond surface-level correlation
  • Built and evaluated machine learning models for predictive analysis
  • Used interpretability techniques to explain model behavior and improve trust in results
  • Organized the workflow into reusable scripts, notebooks, tests, and app-based outputs

📊 Product Metrics & Impact

  • Defined KPIs to measure performance and analytical outcomes
  • Identified trends and behavioral patterns in the data
  • Highlighted opportunities to improve decision quality through evidence-based analysis
  • Enabled data-driven decision-making through interpretable insights
  • Demonstrated how causal thinking can strengthen prioritization and reduce reliance on assumptions

💼 Business Impact

  • Helps teams make more confident, data-informed decisions
  • Bridges the gap between correlation analysis and deeper analytical reasoning
  • Supports prioritization by identifying meaningful drivers of outcomes
  • Demonstrates an end-to-end analytics workflow valuable for product, growth, and strategy teams
  • Shows how analysis can move from reporting to decision support

🛠️ Tools & Technologies

  • Python
  • Pandas
  • NumPy
  • Scikit-learn
  • SHAP
  • Streamlit
  • Pytest
  • GitHub Actions
  • DVC

🚀 How to Run Locally

Clone the repository
git clone https://github.com/Denis0242/CareFlow.git

Navigate into the project
cd CareFlow

Create a virtual environment
python -m venv .venv

Activate the environment on Windows
.venv\Scripts\activate

Install dependencies
pip install -r requirements.txt

Run the application
streamlit run app.py


📁 Project Structure

CareFlow/

├── .devcontainer/ — Development container configuration
├── .dvc/ — Data version control configuration
├── .github/workflows/ — CI/CD workflows
├── data/ — Dataset files
├── notebooks/ — Exploratory and analytical notebooks
├── scripts/ — Reusable analysis scripts
├── tests/ — Automated tests
├── app.py — Main Streamlit application
├── main.py — Project entry point
├── pyproject.toml — Project configuration
├── requirements.txt — Python dependencies
├── setup.py — Package setup
└── README.md — Project documentation


🔍 Use Case

This project demonstrates:

  • Strong analytical thinking beyond basic reporting
  • Experience with causal inference, experimentation logic, and predictive modeling
  • Ability to connect technical analysis to business and product decisions
  • Proficiency in Python-based analytics workflows
  • Ability to communicate insights through an interactive application
  • End-to-end project ownership from analysis to presentation

⭐ Why This Project Stands Out

Unlike a standard dashboard or prediction project, CareFlow is built to show decision-driven analytics. It combines exploratory analysis, machine learning, interpretability, and causal reasoning in one project, making it especially relevant for Product Data Analyst, Data Analyst, and Product Analytics roles.


📌 Author

Denis Agyapong
Product Data Analyst | Data Analyst

About

Product Data Science project demonstrating causal inference, experimentation analysis, interpretable machine learning, and decision-driven analytics for product and growth teams.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors