Benchmark Engineer, AI Safety & LLM Evaluation

Aisafety; Paris; Hybride; EUR 80 000 - 110 000; Plein temps

Avantages offerts par ce poste

Relocation package

Hybrid work (Paris)

Comprehensive medical insurance

All required hardware/tools provided

Résumé du poste

White Circle is seeking a research engineer to build and maintain an internal benchmark suite for single- and multi-turn content and agentic safety. You’ll work with the team on projects studying agent behaviours in the wild and help evolve evals for new features and data.

The role involves Python development, production-grade benchmarks, and collaboration with the product team to ensure robust evaluation of White Circle’s AI safety stack.

Qualifications

• Experience writing Python and building ML benchmarks or eval suites.

Responsabilités

• Own and maintain the internal benchmark suite for single/multi-turn content and agentic safety.
• Build benchmarks that distinguish model capabilities and support evals for flagship models.
• Collaborate with product to cover core functionality and new features in benchmarks.
• Adapt evals to new data and verticals and study real-world agent behaviours.

Description du poste

White Circle is seeking a research engineer to build and maintain an internal benchmark suite for single- and multi-turn content and agentic safety. You’ll work with the team on projects studying agent behaviours in the wild and help evolve evals for new features and data.

The role involves Python development, production-grade benchmarks, and collaboration with the product team to ensure robust evaluation of White Circle’s AI safety stack.