Building the security layer for AI agents, LLM applications, and multi-agent systems. Active open-source contributor across AI Security, MCP/Agentic Security, DevSecOps, and Multi-Agent Systems.
- AI Security & Red-Teaming — LLM vulnerability assessment, jailbreak detection, probe development
- MCP / Agentic Security — Model Context Protocol security scanning, credential detection, tool poisoning
- DevSecOps — SAST rule development, dependency hardening, CI/CD security
- Multi-Agent Systems — Secure agent orchestration, guardrails, observability
Actively contributing to production security and AI tooling:
| Repository | Contribution | Status |
|---|---|---|
| NVIDIA/garak | Transient HTTP error retry (408/429/502/503/504) in OpenAI-compatible generators | PR #1973 |
| NVIDIA/garak | AART probe for AI-assisted red-teaming across diverse cultural contexts | PR #1974 |
| PyCQA/bandit | B905 plugin: Flask send_file() path traversal detection (DevSecOps) |
PR #1459 |
| mcp-safeguard | Replicate + Cohere API token detection rules (CRED-027, CRED-028) | PR #8 |
| google/adk-python | OpenTelemetry dependency compatibility update | PR #6426 |
| guardrails-ai/guardrails | Modernized Python typing annotations across codebase | PR #1581 |
| agentic-radar | Rich dependency bump for agentic security scanner | PR #131 |
Languages: Python (primary), C++, JavaScript, TypeScript, SQL
AI/ML: PyTorch, TensorFlow, scikit-learn, NumPy, Pandas, LangChain, LangGraph
Security: Garak, Bandit, Semgrep, MCP security scanning, OWASP Top 10 for LLM & Agentic AI
Infrastructure: Docker, Git/GitHub, FastAPI, MongoDB, MySQL, CI/CD, GitHub Actions
Open to collaboration on AI security tooling, MCP security research, and multi-agent system hardening.