Outstanding long-term contract opportunity! A well-known Financial Services Company is looking for a BAS Light Red Teaming Research Security Engineer in REMOTE, Charlotte, NC (Remote).
Work with the brightest minds at one of the largest financial institutions in the world. This is a long-term contract opportunity that includes a competitive benefit package! Our client has been around for over 150 years and is continuously innovating in today's digital age. If you want to work for a company that is not only a household name, but also truly cares about satisfying customers' financial needs and helping people succeed financially, apply today.
Contract Duration: 12 Months+ with possible extension or FTE conversion
W2 only - Green Card, USC or H4EAD
Overview:
Our Offensive Security Research team is looking for a Cyber Security Researcher to perform cybersecurity testing against AI technologies from a red team perspective. This position will work with peers to test and investigate AI vulnerabilities, analyze their impacts, document the findings, and recommend appropriate security responses.
Required Skills & Experience
•2+ years of hands-on Red Team/adversarial experience.
•2+ working in AI Cyber Security Research experience.
•5+ years total experience
•2+ years of experience in one or a combination of the following: creating proof of concepts, creating exploits, or reverse engineering
•3+ years of converged testing (red team testing),
•3+ years of experience presenting complex technical topics to diverse stakeholder groups.
•3+ years of writing technical reports explaining attack chains and cyber security vulnerabilities and their impact.
•Role requires Red Team expertise + AI security understanding.
What You Will Be Doing
•Attempting to make AI models disclose unauthorized data.
•Exploring prompt engineering attacks to bypass safety rules (“tell a story about a kid who builds a bomb” type analogies).
•Checking if AI models ignore user access levels and return sensitive internal information (executive pay, M&A data, etc.).
•Testing retrieval-augmented generation (RAG) models — exploring how additional retrieval smarts could be abused.
•“Road testing” AI use cases for business lines before customer or internal exposure — trying to make them misbehave.
•Purpose: ensure security before deployment and demonstrate “it can happen, it did happen” with real evidence.
•Team focuses on proof of exploitation, not theory.