BuildingAI Security

Adversarial Phishing Detection

Simulated arms-race where an LLM attacker and classifier defender co-evolve.

🎯The Problem

Phishing attacks evolve faster than static classifiers can adapt. Real-world security requires adaptive defenses.

💡The Solution

Attacker agent (PhishGen) generates evolving phishing emails. Defender agent (PhishShield) retrains to resist them. Loop controller coordinates the adversarial training.

✨Key Highlights

  • LLM-powered phishing generation
  • Continuous adversarial training
  • Evolution timeline visualization

Other Projects

HyperSentry

An agentic security copilot that uses policy checks + retrieval to generate remediation PRs and accelerate incident analysis.

SchemaPulse

OpenAI-compatible inference layer for schema-valid structured outputs with streaming tool-call parsing.

EdgeJury

Multi-LLM council with cross-review for truthful QA on serverless edge inference.