Home

Resume

Jobs

My applications

Requests

Education

Career advice

Companies

Salary Calculator


Terms & Conditions

Privacy Policy

© 2026 All Rights Reserved

Senior AI Security Engineer

●

12 hours ago

Greece, Attiki, Athens
Full Time
Senior
Remote

Basic Information

Humanbound is a Greek AI security company helping organizations design, test, and operate trustworthy AI systems. Our platform provides automated, large-scale testing and continuous assurance across an organization's AI agents, the ones it builds and the ones its people use, identifying risks such as prompt injection, boundary violations, and emergent behaviors before they reach production, with customized threat coverage for key accounts. We are hiring our first Senior AI Security Engineer, reporting to the CTO.

You own threat intelligence for the product: what the platform tests for, against which agent architectures and platforms, under which regulatory expectations, and how a finding is judged. As the team grows, this position leads the platform's threat intelligence function. For key accounts you also run the engagement: threat model, assessment, protection, and the technical relationship afterwards. This is where your intelligence gets tested against real systems and where the next round of coverage comes from.

Responsibilities

  • Own the attack catalog and taxonomy: what the platform tests for, how findings are classified, and how coverage maps to OWASP Top 10 for LLM and Agentic Applications, MITRE ATLAS, NIST, and the EU AI Act.
  • Keep coverage current: new attack classes, new agent protocols and builder platforms, new regulatory expectations, and what appears in the field.
  • Define domain and regulatory profiles so key accounts are each tested against what applies to them.
  • Represent Humanbound in standards work and publish the first vulnerability disclosures that come out of our research.
  • Run the engagement end to end for key accounts: threat model, assessment, protection, deliver a report the client can act on.
  • Stay the technical contact for those accounts after delivery, and bring what each one teaches back into the product.

Requirements

AI Solutions Design
Artificial Intelligence
Consulting
Cybersecurity
Penetration Testing
Python
Research
SIEM
  • 5+ years in security engineering, application security, or offensive security
  • Able to threat model a system you have not seen before: work out where trust changes hands, decide which threats matter, and say what should be tested.
  • Familiar with agentic threat vectors such as prompt injection, tool abuse, retrieval poisoning, and data exfiltration, and with how agents are built: tool calling, RAG, multi-agent setups, MCP.
  • Working knowledge of OWASP Top 10 for LLM and Agentic Applications and MITRE ATLAS.
  • Awareness of the EU AI Act, DORA, and NIS2, and an interest in how regulation becomes something you can test.
  • Proficient in Python, comfortable with containers and one major cloud (preferably Azure).
  • Experience delivering security assessments and reports to clients.
  • Fluent in Greek and English.

Benefits

  • Competitive compensation package based on experience
  • Remote work flexibility
  • Full equipment provided
  • Stock options
  • Continuous learning & training opportunities
  • Career growth in a fast-paced, scaling environment
  • Flexible working hours
  • Collaborative, dynamic team culture
  • Access to modern tools & technologies

Good to have

  • Published vulnerability disclosures, security research.
  • Experience with adversarial testing tooling such as PyRIT, promptfoo, or garak, or with a harness of your own.
  • Contributions to open-source security tooling.
  • Experience building agents on Copilot Studio, the OpenAI platform, or comparable agent builders.
  • Runtime guardrail, AI firewall, or detection engineering experience.
About the company

-

About the company

-

Humanbound

Humanbound

Information Technology
Security Engineer