Pinned Loading

1. MeshGuardEval MeshGuardEval Public

   MeshGuardEval is a contract‑driven evaluation harness for AI systems, combining QA, security testing, and AI safety. It validates multi‑agent workflows, unsafe prompts, tool‑call behavior, and summ…

2. MeshGuardEval-Multi-Architecture-Evaluation-Model MeshGuardEval-Multi-Architecture-Evaluation-Model Public

   A lightweight, open-source architecture‑agnostic framework for evaluating LLM systems using a simple three‑layer model: Foundation, Interaction, and Outcomes. Possibly in future might include mocke…

3. ML-Plus-JSON-Classifier-Integrated-With-PromptInjector ML-Plus-JSON-Classifier-Integrated-With-PromptInjector Public

   Integrated with this project: https://github.com/nayangoel/PromptInjector

4. VHFBWHCH-Streamlit-App VHFBWHCH-Streamlit-App Public

   A Streamlit interface that loads SQuAD v2 data directly from the Hugging Face API and lets users browse, filter, and explore answerable vs unanswerable question–context pairs in real time.

   Python

5. Learning-React-HTML-Widget-Dashboard Learning-React-HTML-Widget-Dashboard Public

   Using pre built scripts to showcase what I have learned from a Machine Learning course - Logistic Regression

   Python

6. Unit-9-Website-Application-Security-Pentest Unit-9-Website-Application-Security-Pentest Public

   Objective of this assignment was to find vulnerabilities in a website by using common attack techniques: SQL Injection, XSS and Session Hijacking and Fixation.