AI Engineer (Security Research) at Escape Technologies

Paris, France

true · full_time · Visa sponsorship

Apply for this role at Escape Technologies

Responsibilities:
- Conduct original research probing frontier LLMs for offensive and defensive security tasks; design custom benchmarks and experiments.
- Invent and prototype defenses against LLM-powered attackers (adversarial inputs, traps, deception primitives) and evaluate robustness.
- Build reproducible PoCs, tooling and datasets; open-source components and publish papers or reports to establish leadership.
- Collaborate with engineering, product and founders to integrate research outputs into product features (AI pentesting, ASM, DAST).
- Mentor peers on methods, ensure experiments are rigorous, and contribute to threat models and red-team exercises.

Requirements:
- Graduate degree (MSc/PhD) in ML, CS, Math or equivalent research experience, plus 3+ years in ML/security research or offensive security roles.
- Strong hands-on ML skills (LLMs, PyTorch or similar), with experience designing benchmarks, evaluation metrics and adversarial attacks.
- Demonstrated public track record (papers, CVEs, tools, or widely used OSS) showing original contributions to security or ML research.
- Practical offensive security knowledge (pen testing, fuzzing, red teaming) and ability to reason about attacker/defender tradeoffs.
- Excellent communication for publishing and cross-team collaboration; willing to work hybrid in Paris with occasional NYC travel.