ML Platform Architect & Production Infra Lead

DeepHealth; Paris; Hybrid; EUR 50,000 - 80,000; Full time

Job summary

DeepHealth in Paris, Île-de-France, is seeking a Senior Machine Learning Engineer to own the ML Monorepo used by the X-ray, CT, and MRI teams. You will define architecture, tooling, and coding standards for a distributed team of about twenty engineers and scientists, guiding experiments toward production.

You will oversee datasets, training/evaluation loops, model export, and inference servers, ensuring traceability and compliance while boosting performance and developer experience.

Qualifications

• Master's degree or equivalent practical experience.
• 5+ years of software engineering with ML infra/tooling.
• Strong Python and PyTorch; API design and testing.
• Cloud experience with GCP and containerization.
• Productionization: packaging, inference serving, performance constraints.
• English communication in a distributed team.

Responsibilities

• Own architecture and roadmap of the ML Monorepo.
• Build and run engineering tooling for the codebase.
• Own path from trained checkpoint to deployable model.
• Define engineering standards to accelerate research.
• Ensure traceability of data, code, and models.
• Investigate performance and reliability issues.
• Define AI coding agents tooling and guardrails.
• Review code and mentor colleagues.

Skills

Python programming

PyTorch

Cloud experience (GCP)

Containerization

API design

ML platform tooling

English communication

Team collaboration

Model deployment

Education

Master's degree or equivalent practical experience

Tools

Docker

Kubernetes

Git

Slurm

Job description

DeepHealth in Paris, Île-de-France, is seeking a Senior Machine Learning Engineer to own the ML Monorepo used by the X-ray, CT, and MRI teams. You will define architecture, tooling, and coding standards for a distributed team of about twenty engineers and scientists, guiding experiments toward production.

You will oversee datasets, training/evaluation loops, model export, and inference servers, ensuring traceability and compliance while boosting performance and developer experience.