Building intelligent AI systems with clean architecture, reproducible workflows, and scalable machine learning solutions.
I'm a Machine Learning & Deep Learning Engineer passionate about designing intelligent systems with a strong emphasis on clean architecture, modular codebases, and reproducible ML workflows.
My primary interests include Artificial Neural Networks (ANNs), Convolutional Neural Networks (CNNs), computer vision, predictive modeling, and building maintainable end-to-end machine learning pipelines.
- π§© Self-taught ML/DL practitioner
- βοΈ Modular & configuration-driven development
- π Data-driven experimentation and evaluation
- π Continuous learning and open-source development
Machine learning pipeline for fraud detection using Logistic Regression, Random Forest, and XGBoost, featuring SMOTE, cross-validation, and a deployed Streamlit application.
π https://github.com/ArianJr/credit-card-fraud-detection
Comparison of CNN and transfer learning architectures including ResNet50, MobileNetV2, and EfficientNetB0 for image classification.
π https://github.com/ArianJr/garbage-classification-transfer-learning
Deep learning pipeline for classifying chest X-rays into COVID-19, pneumonia, and normal categories.
π https://github.com/ArianJr/chest-xray-covid19-classification
LSTM-based neural network for forecasting Google stock prices with preprocessing, training, and visualization.
π https://github.com/ArianJr/google-stock-price-forecasting-lstm
End-to-end deep learning workflow for regression and classification on academic performance data.
π https://github.com/ArianJr/student-performance-deep-learning
Machine learning classifiers for predicting lung cancer risk using structured clinical data.
π https://github.com/ArianJr/lung-cancer-prediction-ml
π Portfolio
https://arianjr.github.io
πΌ LinkedIn
https://www.linkedin.com/in/arian-jafar
π§ Email
arianjafar59@gmail.com
"Code. Learn. Iterate. Intelligence is built, one model at a time."

