Junior AI Engineer – AI Security
Enregistrez cette offre et organisez votre recherche
Créez un compte gratuit pour enregistrer des offres d'emploi, créer des alertes et revenir à cette liste depuis votre tableau de bord.
En continuant, vous acceptez nos Conditions d’utilisation & Politique de confidentialité.
About Preemptics
Preemptics is an AI security startup. We red-team AI agents: we test how they can be manipulated or misused, so organizations can fix weaknesses before real attackers find them. We're a small, technical team working where AI and cybersecurity meet.
The role
We're looking for a junior engineer who wants to build a career in AI security. You'll work across three areas: red-teaming AI agents, building realistic AI applications to test against, and contributing to the platform that runs our tests.
You'll be mentored by our engineers and red teamers, and you'll get real ownership of the work you ship.
What you'll do
Design and run red-teaming tests against AI agents and LLM applications.
Turn attack techniques into automated, reusable tools in Python.
Build realistic AI agents and applications (with tools, memory and RAG) to use as test targets.
Contribute features, integrations and tests to our testing platform.
Analyze results and document findings clearly.
Follow new AI security research and share what you learn with the team.
What we're looking for
An engineering degree or Master's in computer science, software engineering, cybersecurity, AI or a related field.
Good Python skills.
Hands-on experience with security tools or practices, such as Burp Suite, Nmap, vulnerability scanning or CTF challenges.
An understanding of how LLM applications work (prompts, APIs, tool calling, RAG), or a strong drive to learn it fast.
Knowledge of, or strong interest in, attacks on AI systems: prompt injection, jailbreaks, data exfiltration and tool misuse.
Curiosity, autonomy, and comfort reading and writing technical English.
Eligibility
This role starts with a CIVP contract, so you must be eligible for it: Tunisian nationality, a higher-education degree, and a first-time job seeker (no more than 24 months of continuous or 36 months of non-continuous work since your last degree).
Nice to have
CTF, bug bounty or personal security projects you can show us.
Familiarity with OWASP Top 10 for LLM Applications or MITRE ATLAS.
Experience with an agent framework such as LangChain or LangGraph.
Docker, Git and basic Linux skills.
A final-year project or research in AI or security.
What we offer
A paid CIVP contract to start, with conversion to a permanent CDI.
Hands-on work on real AI security problems from day one.
Mentoring from experienced engineers and red teamers.
Hybrid work in Tunis.
Hiring process
Application
HR interview
Take-home challenge
Technical interview, including a discussion of your challenge
Offer