I am an AI Red Team Lead specializing in the extreme edge of AI security. My expertise lies in proactive threat mitigation—identifying and exploiting critical vulnerabilities in Large Language Models through adversarial attacks, hijacking, and malicious injection testing.

I focus on breaking systems before the concept of "what if" is even considered.

My background is uniquely multidisciplinary, combining a Master’s in Biochemistry from Duke University (USA) with extensive experience in high-stakes AI/ML security operations, typically conducted under strict NDA.

Currently, I am on a professional sabbatical to focus on family (Wife + Mom) while concurrently pursuing foundational Computer Science studies in Stockholm. This GitHub documents my independent research, tinkering, and ongoing projects.

My work involves stress-testing advanced AI systems under extreme scenarios to secure them prior to deployment.

• AI Red Teaming: Leading specialized teams to simulate sophisticated threats against AI/ML infrastructure.
• LLM Security: Deep expertise in prompt injection, hijacking, adversarial inputs, and testing model alignment/safety protocols.
• Threat Modeling: Analyzing and documenting complex attack vectors and extreme-edge misuse cases.
• Interdisciplinary Insight: Applying analytical rigor from a biochemistry background to complex computational and systemic challenges.

Due to the confidential nature of my professional work, most projects are protected under NDAs.

• FayeBlade CLI (Private): My current primary development project; planned for a future public release.
• NSA SIMP: A public repository I actively follow and respect.

I maintain stringent policies regarding professional engagements to ensure absolute legal compliance and prevent conflicts of interest.

I am not currently seeking employment or large-scale contracts.

I am happy to consult on small, specific tasks (academic or open-source), provided there is zero conflict with my existing confidentiality agreements. I maintain a zero NDA conflict state.

For engagements requiring proprietary vulnerability testing ("breaking your toys"), the following prerequisites are non-negotiable. There are zero exemptions.

1. Legal Authorization: Comprehensive legal documentation is mandatory. All documents must be formally notarized, witnessed, and signed by authorized personnel (Owner/Board level).
2. Stakeholder Verification: A direct meeting (in-person or VOIP) with the legal owners or board members is required.
3. Confidentiality & Disclosure: Engagements require strict, mutual NDAs. The Public Interest Disclosure Act (PIDA) must be explicitly defined within the agreement.
4. Ethical & Legal Boundaries: All work must comply strictly with EU, Swedish, and international law. I am willing to develop inert PoCs provided I am granted full, authorized access to the relevant databases/systems.
5. Government Restrictions: I do not accept contracts from government agencies (foreign or domestic), with the sole exception of explicitly tasked CSAM (Child Sexual Abuse Material) prevention initiatives.

While my professional life revolves around security analysis and tech innovation (Google Gemini fangirl status in full effect), my roles as a Mom and Wife are paramount. I am a lifelong learner dedicated to the success of the next generation and proud of my European heritage.

• Interests: PC Gaming (Master Race).
• Outdoors: Hiking, hunting adventures, skydiving, and firearms proficiency (🇸🇪🇺🇸).