Xiangxin Guardrails is an open-source, context-aware AI guardrails platform that provides protection against prompt injection attacks, content safety risks, and data leakage. It can be deployed as a security gateway or integrated via API, offering enterprise-grade, fully private deployment options.

OpenGuardrails: A Configurable, Unified, and Scalable Guardrails Platform for Large Language Models

The one-stop DLP browser extension to stop users from sharing sensitive information with chatGPT.

A GUI Based tool developed in Rust & Electron for preventing Sensitive Information Leaks

A straight forward Data Leakage Protection solution for Apache / NGINX that intelligently blocks access to the most sensitive data types. Offering Zero day protection for the top 10 data hacking targets across your Apache/AWS/RDS/Mysql and web based services.

The official Python library for the OpenGuardrails API

Vigilante Vixen has learned that there were many security vulnerabilities from their technical, behavioral, law, and human resources aspects. Despite us not being directly involved in offshore financial services or the legal profession, technology roles have a considerable amount of opportunity to review this case and implement security regulations

School Project for DLP class

An end-to-end machine learning pipeline for detecting fraudulent credit card transactions using advanced resampling techniques and ensemble modeling strategies.

KNN in Forensic Science: Classifying Glass Evidence for Criminal Investigations

This project highlights the dangers of data leakage in machine learning, showcasing how it can lead to misleadingly high model accuracy, and emphasizes the importance of rigorous validation and awareness to ensure realistic predictions.

"An automation tool designed to help users report data broker websites that display and sell personal information. By collectively reporting these sites to Google Safe Browsing, we can help protect our privacy without relying on costly services that can still be prone to leaks."

Regression | Analysis | Modelling | Bayesian Search CV | Data Leakage | Overfit/Underfit | End-to-End Project

The one-stop DLP browser extension to stop users from sharing sensitive information with chatGPT.