AI Red Teaming Specialist

About the Astek Group

Founded in France in 1988, Astek Group is a global partner in engineering and IT consulting. Leveraging deep expertise across a wide range of industrial and technological sectors, Astek supports international clients in the development and delivery of products and services, while actively contributing to their digital transformation initiatives.

Since its inception, Astek Group has built its growth on a strong culture of entrepreneurship and innovation, as well as on the continuous development of the skills of its more than 10,000 employees, who work every day on diverse and challenging engineering and technology projects.

Join a rapidly growing group in France and worldwide, with 2024 revenues of €705 million.

For more information, please visit: https://astek.net

Role Overview

We are seeking a proactive and technically skilled AI Red Teaming Specialist to join a security-focused AI/ML project. In this role, you will be responsible for evaluating the security, safety, and resilience of AI systems, with a strong focus on generative, reasoning, and agentic Large Language Models (LLMs).

You will adopt an adversarial mindset to identify vulnerabilities and failure modes in AI-driven applications and provide actionable recommendations to harden systems against real-world threats before production deployment.

Project Context

The project is part of an advanced AI security initiative focused on secure, confidential, and resilient AI solutions. The work spans infrastructure, AI workloads, and application-level security, with a strong emphasis on agentic AI systems and next-generation AI threat models.

The environment is highly technical, research-driven, and collaborative, involving close interaction with engineering, data science, and product teams.

Key Responsibilities

• Conduct AI Red Teaming Exercises

  • Plan and execute end-to-end red teaming operations and adversarial simulations targeting LLM-powered systems and applications

• Design & Execute Adversarial Attacks

  • Develop and perform attacks such as:

    • prompt injection

    • data poisoning

    • jailbreaking

    • model evasion and misuse scenarios

  • Identify weaknesses, unsafe behaviors, and failure modes in generative and agentic AI systems

• Vulnerability Analysis & Reporting

  • Systematically document findings and analyze red teaming results

  • Produce clear, high-quality reports describing:

    • identified risks and vulnerabilities

    • impact and severity

    • actionable remediation recommendations

  • Communicate results to both technical and non-technical stakeholders

• Tooling & Automation

  • Contribute to the development and improvement of internal tools and frameworks for AI security testing

  • Work with automated prompt generation and scenario testing tools such as:

    • Garak

    • PyRIT

    • custom red teaming solutions

• Cross-Team Collaboration

  • Work closely with data science, engineering, and product teams

  • Ensure AI security considerations are embedded throughout the AI development lifecycle

  • Provide expert guidance on AI security best practices

• Research & Threat Intelligence

  • Continuously research emerging AI security threats, adversarial ML techniques, and evolving attack vectors

  • Stay current with industry trends, academic research, and real-world incidents

Technical Requirements (Must Have / Preferred)

• Strong experience in AI security, adversarial ML, or offensive security roles

• Hands-on experience red teaming LLMs or generative AI systems

• Familiarity with:

  • OWASP Top 10 for LLMs

  • NIST AI Risk Management Framework

  • AI guardrails systems (e.g. Amazon Bedrock Guardrails, NVIDIA NeMo Guardrails)

• Experience with cloud platforms:

  • AWS, GCP, or Azure

• Understanding of MLOps pipelines and AI deployment workflows

• Preferred certifications:

  • Offensive Security Certified Professional (OSCP)

  • Certified AI Red Teaming Professional (CAIRTP)

• Background in one or more of the following is a plus:

  • content moderation

  • disinformation analysis

  • cyber-threat intelligence

Soft Skills & Availability

• Strong analytical and critical thinking skills

• Clear communication and reporting abilities

• Ability to work with distributed, international teams

• Working hours requirement:

  • overlap with San Francisco timezone

  • either:

    • 4 days per week until 7:00 PM CET, or

    • 2 days per week until 9:00 PM CET

What We Offer

• Long-term collaboration – stability and ongoing career opportunities

• Technical training and certifications – continuous skill development and professional growth

• Mentoring through our Competence Center – from day one, become part of a community that allows you to enhance your skills, participate in conferences, and share knowledge with colleagues facing similar challenges

• Clear career path – transparent progression opportunities

• Employee benefits package, including:

  • Multisport card

  • Private medical care

  • Life insurance

  • Subsidy for public transport

• Friendly work environment – team-building events, social gatherings, and corporate parties

Referral Program

Do you know someone who might be interested in this offer? Take advantage of our referral program and earn a bonus of up to PLN 7,000!

Link: https://astek.pl/system-rekomendacji/

Privacy Notice

The administrator of your personal data is ASTEK Polska sp. z o.o., located in Warsaw (00-133) at Al. Jana Pawła II 22. You have the right to access your data, request its deletion, and other rights regarding personal data. Detailed information on data processing can be found here: https://astek.pl/polityka-prywatnosci

You may withdraw your consent at any time. To withdraw consent, please contact us via email at privacy@astek.pl or by writing to the data administrator at the address above.

AO217841