Platform Engineer for Orchestral: Scale AI Training

Mistral; Paris; Presencial; EUR 90.000 - 120.000; Jornada completa

Descripción de la vacante

Mistral is seeking a Research Engineer for the Orchestral team to help build the agentic platform used to train and evaluate models at scale. You will own core components end-to-end, from design to production, in a distributed system focused on resiliency and reliable workloads.

The role emphasizes building robust tooling for researchers, improving scheduling and routing, and elevating code quality through testing, reviews and CI/CD.

Formación

• Experience building and debugging distributed systems at scale.
• Strong Python skills and software-design instincts: testing, code review, CI/CD.
• Builder mindset with systems built from the ground up.
• Track record in small teams running complex systems at scale (finance, crypto, ad tech).
• Self-starter, low-ego, collaborative with a product mindset toward users.
• Curiosity about ML and RL; willingness to learn.

Responsabilidades

• Design and build the core of the Orchestral platform: library, abstractions and tooling.
• Evolve the rollout engine with capabilities like long-running tasks and subagents.
• Run Orchestral as a production system; keep workloads reliable at scale.
• Improve data and job flow: scheduling, routing and request handling.
• Treat researchers as users and translate recurring needs into robust tooling.
• Raise engineering quality across the platform: testing, code review, CI/CD, observability and docs.
• Contribute to open-source rollout-engine and agent-infrastructure projects where it benefits the platform.

Conocimientos

Python

Distributed systems

Builder mindset

Small teams

Self-starter

ML/RL curiosity

Open-source contributions

Herramientas

Ray

Descripción del empleo

Mistral is seeking a Research Engineer for the Orchestral team to help build the agentic platform used to train and evaluate models at scale. You will own core components end-to-end, from design to production, in a distributed system focused on resiliency and reliable workloads.

The role emphasizes building robust tooling for researchers, improving scheduling and routing, and elevating code quality through testing, reviews and CI/CD.