NeLux: High-performance video processing for Python, powered by FFmpeg and PyTorch. Ultra-fast video decoding directly to tensors.

Screw type detection using ESP-EYE, YOLOv5, and TensorFlow Lite Micro for real-time classification on ESP32.

This repository contains a collection of simple yet insightful Data Science projects that showcase my ability to apply analytical thinking, build ML models, and visualize results.

Decision Tree are the typically machine learning algotithm that are work on specific data and generate the tree with yes or no formate.

Open-sourced is_even function powered by LLM

Real-time cardiovascular risk prediction platform built with Next.js, Flask, and a robust ML pipeline.

AI/ML Engineer Portfolio | End-to-end projects (Customer Churn Streamlit, Credit Card Fraud Detection with XGBoost+Docker+SHAP, Japanese Sentiment Analysis with BERT) | Production-ready (Streamlit Cloud + Docker) | Targeting mid-level roles in Japan

End-to-end ML project predicting household energy consumption | EDA → Preprocessing → XGBoost | R² 89.4% | Python, scikit-learn, XGBoost

morphogenetic language systems

Forging a cutting-edge Salary Prediction Software for Software Engineers. Seamlessly blending the art of data science and the precision of machine learning.

REMX API is an automated REM (Random Encounter Model) tool for animal abundance estimation using Machine Learning

eal-time financial dashboard using Random Forest to predict 5-minute market movements. Built with Python, Scikit-Learn, and Streamlit.

AI-Powered Medical Image Diagnosis Platform | FastAPI, Next.js, MobileNetV2, Secure Reports, HIPAA-Compliant

Rank 1st in Public Score with 0.447021 F1-Macro

AI-based system that predicts patient health risk using machine learning and provides precaution suggestions.

✨ Special repository: My professional Data Science portfolio and AI journey.

Flight Price Prediction

Resume–Job Description Fit Analyzer using NLP, TF-IDF + Logistic Regression and Sentence-BERT with a Streamlit web application for semantic similarity analysis.

Machine-learning dashboard to predict whether a cell cluster is benign or malignant using Python and Streamlit.

An interactive Power BI dashboard analyzing car sales data with filters for location, ship mode, and year. Provides insights into sales trends, budget performance, and profitability with key KPIs and visual breakdowns by segment, category, and month.