Autonomous Pentesting Engineer

About the role

The Autonomous Pentesting Engineer will design and evolve an autonomous penetration‑testing platform that operates within Devoteam Cyber Trust’s Offensive Engineering and Innovation team. The platform aims to automate large parts of the pentesting lifecycle, integrating agents, large language models and existing security tooling to deliver reliable, real‑world testing outcomes.

Key responsibilities

• Define and continuously improve the architecture of autonomous pentesting agents.
• Build controlled execution pipelines covering tasking, tool usage, feedback loops and state management.
• Integrate and optimise LLM‑based components within agent workflows and validate tool‑calling interfaces.
• Ensure the platform aligns with established methodologies such as the OWASP Web Security Testing Guide.
• Test agents in real penetration‑testing scenarios, analyse failures and implement mitigation strategies.

Required profile

• Strong critical thinking with the ability to operate in imperfect, evolving systems.
• Experimental, outcome‑driven mindset with high autonomy and ownership.
• Focus on control, reliability and predictability of automated security processes.

Required skills

• Proficiency in Python.
• Experience with APIs and distributed systems.
• Practical knowledge of Large Language Models (LLMs) and their integration.
• Understanding of agent‑based system design.
• Skill in task decomposition, planning and state management.
• Capability to orchestrate security tools and validate output evidence.

What we offer

• Valuing and supporting talent development.
• Collaboration in a fast‑growing, evolving company.
• Strong organisational culture based on collaboration, flexibility, integrity and low ego.
• Commitment to equal opportunities and inclusion of people with disabilities.