I'm a Security Engineer focused on AI Security — building production AI integrations and applying offensive-security discipline to LLM-integrated systems. Based in Tampa Bay, FL.
Currently: Security Engineer at Abacode Cybersecurity (penetration testing + SOC background). Pursuing the HTB Certified Offensive AI Expert (COAE) credential. Finishing my M.S. in Cybersecurity at Western Governors University.
Certifications: PNPT (Practical Network Penetration Tester) · CompTIA Security+
- Indirect Prompt Injection in RAG (with Layered Defenses) - Hands-on demo of OWASP LLM01 (Prompt Injection). RAG pipeline poisoning attack with spotlighting + injection-classifier defenses. Maps to OWASP LLM Top 10 and MITRE ATLAS.
- Excessive Agency in Tool-Calling Agents (with Layered Defenses) - Hands-on demo of OWASP LLM06 (Excessive Agency). Tool-calling agent hijacked via indirect prompt injection, with layered defenses. Maps to OWASP LLM Top 10 and MITRE ATLAS.
- System Prompt Leakage in LLM Apps (with Layered Defenses) - Hands-on demo of OWASP LLM07 (System Prompt Leakage). Six extraction techniques against a vulnerable agent, with three-layer architectural defenses (no-secrets-in-prompt, output filtering, refusal patterns). Maps to OWASP LLM Top 10 and MITRE ATLAS.
More AI Security projects in progress. LLM05 (Improper Output Handling) is next.
- Building a SOC + Honeynet in Azure (Live Traffic)
- Setup a SIEM in Microsoft Azure (Sentinel)
- Configuring Active Directory in Azure
- Network Security Groups (NSGs) and Network Protocols
