I build machine learning systems that protect financial decisions with accuracy, clarity, and engineering discipline, models built for the real world, not just a notebook.

About.subtitle
I didn’t start in data by chasing models, I started by trying to understand how systems behave. What began as curiosity about patterns and signals grew into a commitment to building tools that help people make clearer, more confident decisions.
Today, I focus on fraud detection, risk analytics, and the data systems that support them. I care about clean pipelines, reliable features, and models that hold up in production, not just in a notebook. My work blends engineering discipline with a mindset shaped by faith, family, and movement. I believe great systems outperform great individuals, and I bring that philosophy into every collaboration.
Looking ahead, I’m driven by one question: does the system help someone make a better call? That’s the standard I build toward from ingestion to validation, modeling to deployment, and everything in between.
My work focuses on the intersection of Fraud Detection, Risk Analytics, Predictive Systems, and Data Engineering.
Delivered 0.93 Precision and 0.82 Recall on a 0.17% imbalanced dataset - 93% of flagged transactions were confirmed fraud.
Achieved a 51% MAE reduction with a production-ready 12-hour demand forecasting pipeline deployed via Docker and cloud infrastructure.

Delivered a 100% clean data SLA across 1M+ records/day with automated quality checks and sub-4-minute pipeline cycles.
Identified the Tenure Effect - customers in their first 12 months are the highest-risk segment, enabling proactive retention strategies.
Python, SQL, AWS & Machine Learning
Professional Impact & Results
Experience.experienceDesc1
Experience.experienceDesc2
Education.subtitle
Education.educationDesc1
Education.educationDesc2
Education.educationDesc3
Let's turn your data into business decisions
delphin.kaduli@gmail.com
Location
Washington, Dc | Open to relocate