Security engineer and AI security researcher. I design and build enterprise-grade security systems — LLM/agent guardrails, autonomous red-teaming, SAST, agentic identity (IdP), and cloud security — and I break AI systems before attackers do.
🛡️ ISC2 member · 🚩 CTF competitor · 🥇 ISACA phishing champion · 🎓 MS Cybersecurity, Yeshiva University
- 🔭 Building systems for AI/LLM security, secure SDLC, and cloud posture
- 🧠 Researching agentic red-teaming, prompt-injection & tool-abuse defenses
- 💬 Ask me about LLM guardrails, AI red-teaming, SAST, and AWS security
- 📫 Reach me at saivarmadpr@gmail.com
| Domain | What I ship |
|---|---|
| 🛡️ AI & LLM Guardrails | Guardrail frameworks and testing harnesses that keep LLM/agentic apps safe, policy-aligned, and jailbreak-resistant. |
| 🤖 AI Red-Teaming | Agentic red-team systems that autonomously hunt prompt injection, tool-abuse, and jailbreaks across multi-turn attack chains. |
| 🔍 SAST & AppSec | Static analysis and secure-SDLC tooling that catches vulnerabilities before they ship. |
| 🪪 Agentic Identity (IdP) | Identity, authn/z, and least-privilege scoping for autonomous AI agents. |
| ☁️ Cloud Security | Secure AWS architectures, cloud posture management (CSPM), and deception/honeypots that surface attacker TTPs. |
AI / ML & LLM Security
Offensive Security
Cloud & DevSecOps
Languages
🤖 wb-red-team · votal-ai-hq
Whitebox & blackbox red-teaming framework for LLMs & agentic AI. Analyzes app source code to map tools, roles, and guardrails, then auto-generates adaptive multi-turn attack chains across categories to surface real vulnerabilities.
|
A platform for testing and benchmarking LLM guardrails — probing safety boundaries, policy compliance, and failure modes for AI/agentic apps. |
Capstone: AI-deployed honeypots that lure attackers and analyze their TTPs, plus cloud misconfiguration scanning to cut incidents. |
|
Secure AWS reference architecture — VPC, IAM, WAF, GuardDuty, Security Hub, and CloudFormation automation following AWS best practices. |
Writeups from Yeshiva University's CTF — web exploitation, cryptography, and network forensics, with custom scripts and tooling. |
- 🥇 1st place — Phishing Competition, ISACA
- 🥈 2nd place — CTF Competition, Yeshiva University
- 📜 ISC2 Certified in Cybersecurity (CC) · ISC2 Member
- 🎓 MS in Cybersecurity — Yeshiva University
