AI Security Red Team Engineer

DHL; Warszawa; Hybrid; Confidential; Full time

Benefits offered by this job

Flexible remote work up to 3 days/week

Health and preventive care services

Family-friendly supports

Mobility benefits including companybik

Onboarding and development programs

Pension plan and social counseling

Job summary

DHL Group is seeking a Senior AI Security Architect to lead our AI security strategy. You will guide senior stakeholders, conduct AI red team engagements, evaluate AI security products, and develop new tools and TTPs. A strong background in offensive security, AI threats, and communication skills is essential.

The role emphasizes staying ahead of cyber threats, documenting findings, and collaborating with cross-functional teams to align security with DHL Group's dynamic needs.

Qualifications

• At least 5 years of proven experience in offensive security engagements (red teaming, pen testing, vulnerability research, exploit development).
• At least 3 years of proven experience in Web exploitation within red team activities and a year of proven experience in AI red teaming.
• Deep technical understanding of AI threats and TTPs (MITRE ATLAS, OWASP TOP 10 LLM).
• Excellent coding and scripting skills and ability to develop robust custom tools and automation.
• Proven experience in AI vulnerabilities analysis.

Responsibilities

• Be SME on Offensive AI Security and guide senior stakeholders.
• Conduct AI Red Team exercises to uncover vulnerabilities and evaluate security posture.
• Perform AI Security Product Evaluation and test security solutions.
• Develop and uplift AI red teaming tools and new TTPs.
• Perform comprehensive risk assessments and produce actionable reports.
• Maintain detailed documentation of red team activities and present to senior management.
• Stay updated with AI security trends and cyber-threats.
• Communicate effectively with stakeholders.

Skills

Offensive security

Red teaming

Penetration testing

Vulnerability research

Exploit development

AI threats & TTPs

English communication

Scripting

Automation

Education

BSc/MSc in CS or IT

Tools

AI frameworks

AI tool stacks

Python scripting

Job description

DHL Group - we are the world's leading logistics provider. Our company connects people and markets, enabling global trade. We pursue the strategic goals of being the first choice worldwide for customers, employees, and investors. Through responsible entrepreneurial actions and commitment to society and the environment, we make a positive contribution to the world. The Group CISO Architecture team provides a group-wide architecture framework for cybersecurity which includes technical and organizational requirements as well as guidelines for cybersecurity for IT and OT. They are seeking a highly skilled and visionary Senior AI Security Architect (m/f/d) to lead and advance our AI security strategy at the heart of DHL Group. This pivotal role requires a strategic thinker who can prepare and negotiate decisions for the highest technical committees, ensuring that our security framework aligns with the dynamic needs of our organization and the ever-evolving threat landscape.

Your responsibilities

• Be a subject matter expert (SME) on Offensive AI Security – provide guidance and steer for senior Stakeholders, non-technical as well as technical experts.
• AI Red Team Excercises - conduct offensive technical cybersecurity engagements of varied nature, plan and execute long running AI Red teaming testing operations (simulated cyber attacks) against AI systems, AI models and their operating environment to uncover vulnerabilities, evaluate the security posture of the AI system, its models and infrastructure, identifying vulnerabilities and misconfigurations.
• Perform AI Security Product Evaluation – perform offensive security testing and evaluation of various AI security solutions / products.
• Develop and uplift AI red teaming tools - as well as new TTPs improving testing coverage and effectiveness (e.g. PyRIT, Garak, HexStrike, Giskard, DeepTeam, etc.)
• Perform comprehensive risk assessments - and create reports detailing the findings characteristics (impact, complexity, exploitability, etc.), prioritization and recommendations.
• Record Keeping - maintain detailed documentation of all red team activities, findings, recommendations and present reports to senior management and key stakeholders.
• Stay up to date with the latest trends and development in AI security and cyber-threats.
• Have excellent communication skills – ensure continuous stakeholder management.

Your profile

• At least 5 years of proven experience in offensive security engagements (red teaming, pen testing, vulnerability research, exploit development) preferably conducted multiple TIBER-EU / CBEST or similar structured exercises.
• At least 3 years of proven experience in Web exploitation within red team activities and a year of proven experience in AI red teaming.
• Deep technical understanding of
  • AI threats and TTPs (MITRE ATLAS, OWASP TOP 10 LLM, etc.)
  • AI technologies, AI frameworks, AI technology stacks and tools.
• Excellent coding and scripting skills and ability to develop robust custom tools and automation to support offensive activities.
• Proven experience in AI vulnerabilities analysis
• Excellent communication and presentational skills
• Excellent English language skills.

Nice to Have

• Proven records of vulnerability disclosure (CVEs) or publication in recognized conferences / papers (e.g. Blackhat, DEFCON, C3, etc.)
• Advanced offensive security certifications (OSCP+, OSEE, OSCE3, SEC542, SEC522) are highly desirable.
• AI Red Team certifications (AIRTP+, CAIRTP, CAIPT-RT,C-AI/MLPen) are highly desirable.
• Proven contributions to open source / commercial offensive tools development preferably AI security tools.
• BsC/MsC university degree in computer science, information technology.

Your benefits

• Flexibility: We offer up to 3 days/week of flexible remote work, along with flexible working hours without core hours. The arrangements may vary depending on the task and role.
• Well-Being: Fun, team spirit, and health are crucial factors. Therefore, within the company, we offer a wide range of health and preventive care services, an occupational physician and the FC Deutsche Post. Here, employees engage in sports activities together or exchange ideas.
• Family-friendly: The compatibility of family and work has many facets, such as childcare, care for family members, etc. We support this, for example, through kindergarten places as well as the Betreuungs- und Erholungswerk (support and recreation program).
• Mobility: We move not only letters or parcels but also our employees through offerings such as company bicycles, job tickets and possibly company cars.
• Location: The Post Campus brings together people from more than 60 countries. The Post Tower is easily accessible by car, public transportation and bicycle. To facilitate the workday, the campus offers amenities such as parking spaces, post office, café, canteen and more.
• Onboarding: Individual onboarding, training on the job, networking, a dedicated certified program, etc., lay the foundation for a successful start at DHL Group.
• Development opportunities: We offer numerous development opportunities through a comprehensive training landscape, junior talent programs, coaching etc.
• Social commitment: Through global corporate programs, we improve the educational and career opportunities for young people and engage in disaster management at airports. We also contribute to local environmental protection and aid projects initiated by employees on a voluntary basis.
• What else? Company pension plan, social counseling, inclusion consulting, cleaning service, insurance cooperation, corporate discounts, events, sabbaticals and much more!

About DHL Group

Without the commitment, creativity and know-how of our employees worldwide, the success of DHL Group would not be possible. Get to know us and become part of our team.

At DHL Group, applications from all individuals are welcome, regardless of age, gender, origin, religion or physical attributes.