🚀 Amit Vitekar

AI Security Engineer • LLM Red Teamer • Hardware & Telecom Security Researcher

I break things — LLMs, agents, RAG pipelines, firmware, basebands, RF systems — and help teams secure them.
My work spans AI Red Teaming, Adversarial ML, Agent Security, 5G/RAN fuzzing, and embedded exploitation.

🔒 What I Do

• LLM Red Teaming: jailbreaks, prompt injection, tool-call hijacks, agent misuse
• RAG Security: embedding poisoning, retrieval hijacking, vector-db attacks
• Adversarial ML: backdoors, evasion, poisoning, model extraction
• Hardware/Telecom Security: SDR-based fuzzing, secure-boot bypass, LTE/5G protocol abuse

🧪 Selected Work

• Adversarial LLM Red Teaming Toolkit — Multi-model jailbreak & robustness evaluation
• Damn Vulnerable LLM/Agent Framework — RAG poisoning + agent exploitation testbed
• 5G/RAN Fuzzing Harnesses — AFL++ / LibAFL on GTP, NGAP, PFCP, O-RAN
• Evil-Boot — Secure-boot bypass for embedded devices

🧰 Tools I Use

BITE • garak • Rebuff • HouYi • LangChain • LlamaIndex • FAISS
AFL++ • LibAFL • Scapy • USRP • BladeRF • Python • C • Docker • Linux

🎓 Training

100x Engineers GenAI Cohort, HTB AI Red Teamer, Microsoft AI Red Teaming, NVIDIA Adversarial ML

🤝 Connect

🔗 linkedin.com/in/amitvitekar

🐉 Fun bit:

I break AI systems by day and hardware devices by night.
Few people do both — I enjoy being one of them.