Security Research & Projects
A focused selection of AI security research and projects demonstrating LLM red teaming, secure ML systems, and defensive security work.
BlackHalo↗
AI-powered security platform for automated vulnerability discovery, threat intelligence, and security operations.
Visit site →AI Security & LLM Red Teaming
Security-focused research and tooling for prompt injection attacks, jailbreak discovery, and adversarial ML.
View project →Secure LLM Fine-Tuning
Fine-tuning LLMs with safety guardrails using LoRA adapters, with security-focused evaluations.
View project →Secure RAG Pipeline
Retrieval-augmented generation system with input sanitization and prompt injection defenses.
View project →Transformer Security Analysis
Deep-dive into transformer internals to understand attack surfaces and defense mechanisms.
View project →