StromIA

AI roles

LLM Security Specialist

Secures language-model applications against injection, data leakage and tool abuse.

Cybersecurity and AI SecuritySenior SpecialistTechnical or Team Lead

Core responsibilities

  • Separate instructions, data and actions
  • Design targeted adversarial tests
  • Validate controls when models change

Technical skills

  • Direct and indirect prompt injection
  • Context isolation and per-tool permissions
  • Impact measurement and adversarial validation

Behavioural skills

  • Explain severity with a reproducible scenario
  • Work with the teams responsible for remediation

Typical deliverable

A test plan covering document leakage and permission escalation through an assistant.

Possible assessment methods

Demonstrate indirect injection in an authorised environment and verify layered defences.

In-depth technical assessment

Let’s talk about your next AI hire.

Thirty minutes to understand what you need and tell you how we can help.