End-to-end data engineering pipeline: multi-source ingestion → Postgres warehouse → dbt star schema with data-quality tests → Airflow orchestration, all containerized with Docker.

airflowdata-engineeringdbt

dockeretl

postgresql

deepfake-detection

Repo · 0 stars · Jupyter Notebook

▲ 0

Robust deepfake detection using learning-based CV methods with cross-dataset generalization

analyzing-airbnb-pricing-dynamics-in-newyork-city

Repo · 0 stars · Jupyter Notebook

▲ 0

Data mining project analyzing Airbnb pricing patterns in New York City using EDA, clustering, and regression models.

whatsdule

Repo · 0 stars · JavaScript

▲ 0

WhatsDule is an automation and scheduling platform that integrates WhatsApp messaging, task scheduling services, and workflow orchestration to send automated reminders and notifications, powered by backend APIs, workflow configs, and database integration.

nyc-school-bus-delay-prediction

Repo · 0 stars · Jupyter Notebook

▲ 0

Machine learning system to predict NYC school bus delays using 1.2M+ breakdown & delay records. Includes data cleaning, feature engineering, EDA, XGBoost modeling, and regression/classification pipelines.

deepfake-detection-models

Model · 0 stars

▲ 0

Model · video-classification · 0 downloads · 0 likes

pytorchdeepfake-detectionvideo-classificationensemble

About

Data/ML engineer with 2+ years across production software (Samsung R&D) and data engineering, building ETL/ELT pipelines and deploying ML systems end to end. Core stack: Python, SQL, PySpark, PostgreSQL, Airflow, Docker, AWS, PyTorch. M.S. in Data Science, Pace University; specializing in data engineering and applied AI and machine learning.