Skip to content
View DelphinKdl's full-sized avatar
🎯
Focusing
🎯
Focusing
9 followers · 8 following

Achievements

Achievement: Pull SharkAchievement: Quickdraw

Achievements

Achievement: Pull SharkAchievement: Quickdraw

Block or report DelphinKdl

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
DelphinKdl/README.md

Delphin K.

Data Scientist | Analytical Engineer

Predictive Systems • Revenue Optimization • Business Intelligence

linkedin logo website logo

The Philosophy

I don't just build models. I build answers.

I bridge the gap between technical engineering and business strategy. I specialize in architecting end-to-end systems—from cleaning raw SQL datasets to deploying CatBoost predictive models on AWS that protect ARR and drive measurable ROI.

The Toolkit

Modeling & Analytics Infrastructure & Cloud Visualization & BI
Python CatBoost Optuna SQL AWS (EC2/Lambda) Power BI (DAX)
Scikit-Learn Pandas Docker Git Flask Plotly Seaborn

Core Focus Areas:

  • Predictive Systems: Real-time Fraud Detection & Demand Forecasting (CatBoost, XGBoost).
  • Revenue Optimization: Churn mitigation and ARR protection through stratified modeling.
  • Analytical Engineering: Modular ML pipelines using DataManager and PipelineRunner patterns.
  • Business Intelligence: Transforming high-dimensional data into executive Power BI stories.

Featured Systems

  • 🚲 Demand Forecasting: Cloud-deployed API with a 51% MAE reduction using TimeSeriesSplit.
  • 📉 Churn Prediction: Targeted retention system protecting ARR with 0.64 precision.
  • 🛡️ Fraud Detection: Real-time predictive engine optimized for high-stakes financial security.

Let's Connect

Currently open to Full-time Roles | Washington, DC or Remote

Visit datawithdelphin.com

Pinned Loading

  1. Demand-Forecasting Demand-Forecasting Public

    A Machine Learning Application that forecasts hour-ahead bike rental demand across an entire city, enabling dynamic pricing optimization and revenue maximization.

    Jupyter Notebook 2

  2. Vodafone-customer-churn-ML-prediction Vodafone-customer-churn-ML-prediction Public

    A Machine Learning Application that predicts customer churn for companies, enabling proactive retention strategies and reducing revenue loss.

    Jupyter Notebook

  3. -Data-analysis-project-Data-Driven-Insights-for-Indian-Startup-Growth -Data-analysis-project-Data-Driven-Insights-for-Indian-Startup-Growth Public

    Welcome to the repository for the Indian Startup Funding Analysis project. This data analysis project was commissioned by a newly funded Indian startup. Our primary goal is to provide valuable insi…

    Jupyter Notebook 2

  4. Time-Series-Regression-Analysis Time-Series-Regression-Analysis Public

    This is a time series forecasting problem. In this project, you'll predict store sales on data from Corporation Favorita, a large Ecuadorian-based grocery retailer.

    Jupyter Notebook