Real-time machine learning system forecasting major (M/X class) solar flares within 24 hours using NASA SDO SHARP magnetic parameters and NOAA GOES X-ray data. Benchmarked via True Skill Statistic (TSS = 0.650) with interactive live space-weather dashboard.

End-to-end Predictive Maintenance & KPI Dashboard using Machine Learning (RUL prediction), FastAPI, Streamlit, and SQL.

Command-line analysis and surrogate models for CFD results (OpenFOAM, CSV, VTK)

Comparative tea leaf disease classification using VGG16, LBP, GLCM, HSV, and SVM, evaluated on internal and external image datasets.

Beginner-friendly course on machine learning in R for clinical research (EN/ES). Curso de machine learning en R para investigación clínica, para principiantes (EN/ES). Logistic regression, LASSO, random forest, XGBoost, LightGBM.

🌤️ A beautifully designed, interactive machine learning web app that predicts the weather vibe! Built with Python, Scikit-Learn, and Streamlit featuring a sleek glassmorphism UI.

Automated Pneumonia Detection from Chest X-Rays using InceptionV3 Transfer Learning | 91.83% Accuracy & 98.0% Clinical Recall

面向心理学研究的中文桌面统计与科研构思助手。内置本地计算环境，支持问卷筛选、元分析、中介调节、异质性、网络、DAG、干预方案和机器学习，可配置自己的 AI API。

An enterprise-grade, end-to-end Machine Learning and Cloud Analytics system designed to predict customer churn

An end-to-end machine learning workflow and interactive Streamlit web application that predicts cardiovascular disease risk using a K-Nearest Neighbors (KNN) classifier and clinical patient metrics.

Prove what your AI actually ran.

Train a small language model from scratch on your own computer, and watch it learn to write.

Claude Code plugin that writes evidence-based LaTeX academic project reports from your repository and audits full CRISP-DM coverage (6 phases, 24 generic tasks).

Beginner-to-advanced PyTorch tutorial in 6 levels: tensors, autograd, linear regression, MLP classification, and a ResNet-style CNN pipeline with AdamW, LR scheduling, mixed precision, early stopping, and checkpointing. Runs on synthetic data.

Recursive Web Tensor: GitHub Actions as the distributed control plane for a recursively pruned internet-token learning system

Classifying iris species with Decision Trees and KNN: EDA, cross-validated tuning, precision/recall

This machine learning project uses a Linear Regression algorithm to predict a student's academic marks based on their study hours. It is built in Python using the pandas and scikit-learn libraries.

ANN-based customer churn prediction system built with Python, TensorFlow, Scikit-learn, and Streamlit.

ML-powered Wi-Fi security research platform for anomaly detection, network behavior analysis and authorized wireless security experiments.

Bearing fault diagnosis using vibration signal processing, SVM, STFT, and EfficientNet-B0, with cross-load and cross-fault-size evaluation.

Automated leveraged trading bot for Binance USDS-M futures. Testnet only, not proven profitable (9 hypotheses tested, 9 rejected) - the interesting part is the measurement discipline: ~100 documented root causes, most of them in the layer that measures the strategy rather than the strategy itself.

Predictive Customer Lifetime Value & Revenue Forecasting with LightGBM, SHAP, temporal validation, and purchase propensity modeling.

Native C network anomaly triage: calibrated Isolation Forest, Zeek logs, distribution drift, feature sensitivity, and Rudy Evidence export.

Research platform for 12-lead paper ECG digitization, calibrated PTB-XL multi-label classification, and MIT-BIH single-lead arrhythmia analysis with Groq clinical explanations.

A small GPT that writes molecules as SMILES strings, with an interface to explore them in 2D and 3D. Early version 0.1, more coming.

Projeto de Engenharia de Características para Aprendizagem Computacional - MusicEmotionRecon

IX-AGI: An attempt to build artificial general intelligence through two independent learning systems connected by an adaptive machine corpus callosum. Numerical signals, not AI conversations, coordinate world models and action selection. Julia cognition with native Ada/SPARK execution governance.

Self-learning AI agent that researches the web to build its own skill knowledge bases, then autonomously executes multi-skill tasks with sandboxed code. Built with Python, FastAPI, ChromaDB, and Docker.

Machine learning research on customer segmentation using clustering algorithms, dimensionality reduction, and model evaluation.

Full-stack EV charging load forecasting dashboard: React, FastAPI, XGBoost, TreeSHAP