I build secure AI-powered systems — combining LLMs, penetration testing, and backend engineering to create systems that are intelligent, resilient, and auditable.
┌─ focus areas ──────────────────────────────────────────┐
│ 🔐 LLM Security · prompt injection · agent abuse │
│ 🧪 Secure RAG · retrieval filtering · data bounds │
│ 🌐 OWASP · Web/API pentesting · secure arch │
│ 📡 Detection engineering · SIEM · incident resp │
└────────────────────────────────────────────────────────┘
| Domain | Tools |
|---|---|
| 🤖 AI / LLM | PyTorch · RAG · LangChain · Guardrails · Evaluation |
| 🔐 Security | OWASP · Web/API Pentesting · Threat Modeling · Hardening |
| ⚙️ Backend | FastAPI · Django · PostgreSQL · Secure Auth · Observability |
| 🏗️ Infra | Docker · Linux · CI/CD · Least-Privilege Deployments |
🤖 AI Pentest / LLM Security Model LLM-powered automated pentesting with RAG-assisted context building, structured findings, and security-first orchestration.
PythonLangChainFastAPIRAGDocker
🎙️ Secure AI Voice Assistant FastAPI backend with LLM orchestration, input validation, policy enforcement, and full observability for traces & auditing.
FastAPIPyTorchPostgreSQLOAuth2Docker
🔍 Vulnerability Research Hands-on exploitation & remediation — SQLi, XSS, auth bypass, misconfigurations, and insecure defaults.
Burp SuiteOWASP ZAPSQLMapPython
