LLM Red Teaming
Applying traditional pentesting logic to AI. Identifying jailbreaks, prompt injection vulnerabilities, and data leakage in model outputs.
> pivoting career_vector...
I am a Cybersecurity Expert transitioning into AI Engineering. Applying rigorous defense principles to LLMs, Reinforcement Learning, and Prompt Engineering.
>>> import torch
>>> model.load("llm-v2")
> Starting RLHF...
> Context: Optimized
> Reward Score: 0.92
Years in Security
Prompts Tested
Committed to Protect Customers
Committed to AI Safety
The landscape of technology is shifting, and so am I. After a decade of protecting networks from external threats, I am now focusing on the internal logic of Artificial Intelligence.
I believe that a Security Mindset is the missing link in robust AI development. My goal is to leverage my experience in threat modeling and adversarial thinking to build safer, more aligned AI systems. I am currently deep-diving into the mathematics behind Large Language Models and the practical applications of Post-Training.
Applying traditional pentesting logic to AI. Identifying jailbreaks, prompt injection vulnerabilities, and data leakage in model outputs.
Designing reward functions and collecting human feedback data to fine-tune models for alignment and safety.
Developing robust, production-grade system prompts that maximize model performance while minimizing hallucinations.
Building Retrieval-Augmented Generation (RAG) systems to provide models with secure, up-to-date, and relevant context.
Helping organizations establish policies for ethical AI usage, data privacy, and model deployment standards.
Securing the model supply chain and ensuring integrity during the training and inference phases.
I am passionate about cybersecurity and dedicated to protecting digital assets. As a Cortex Specialist Systems Engineer at Palo Alto Networks, I leverage my expertise in Cortex solutions to design, implement, and support cutting-edge cybersecurity measures for our clients. Key Responsibilities: Solution Design, Implementation Mastery, Technical Expertise, Client Empowerment.
As a Senior Security Technical Specialist at Microsoft, I'm at the forefront of implementing cutting-edge security solutions that drive our clients' success. I'm passionate about helping customers and partners achieve their security and compliance goals through the effective use of Microsoft products. Key Contributions: Strategic Analysis & Implementation, Customer Engagement & Growth, Collaboration & Market Expansion, Product Enhancement, Contribute to Microsoft Security strategy development, Presentations and Demos.
At Cisco Duo Security, I leveraged my expertise in cybersecurity and pre-sales to advocate for and sell the DUO MFA solution as a critical component of Cisco’s Zero Trust solution. My role was pivotal in securing end-user identities against cybersecurity threats and driving the growth of the DUO Multifactor authentication solution. Key Contributions: Product Advocacy & Sales, Cross-Functional Collaboration, Customer Engagement, Product Management Collaboration.
As a Pre-Sales Security Systems Engineer at Cisco, I utilized my deep knowledge of cybersecurity and pre-sales to support our SLED West accounts selling team, product positioning, and selling Cloud solutions. I demonstrated consistent technical leadership and consulting acumen, surpassing FY 20 and FY 21 sales quotas. Key Contributions: Sales Excellence, Technical Leadership, Leadership Development.
As a Network Consulting Engineer at Cisco, I was pivotal in supporting the wireless and mobility teams, managing customer support relationships, and resolving complex customer issues. My unique skills in Lab deployment and Linux knowledge were instrumental in accelerating the team's efforts and delivering successful solutions. Key Contributions: Innovation & Efficiency, Technology Implementation, Customer Retention & Expansion, Innovation Challenge.
I'm passionate about advancing AI Safety, Prompt Engineering, and RLHF while continuing to grow in my current role. If you're working in these areas or exploring how security mindsets can strengthen AI systems, I'd love to connect and exchange ideas.