Applied AI Evaluation Engineer – Benchmark

Il y a 2 jours

Paris, Nouvelle-Aquitaine, France Mistral Temps plein 70 000 € - 110 000 € Contrat

Mistral is seeking an Evaluation Engineer to own end-to-end evals, working with customers, industry experts, and cross‑functional teams to define high‑impact AI use cases. You will design benchmarks, build real-software evals, and push for robust, reusable evaluation frameworks across models and competitors.

You will source data, score outcomes, and present findings to senior leaders, with opportunities to influence go-to-market evidence and future r&d directions.