RL ML Infra Engineer — Hybrid Paris (Relocation)

Npv; Paris; Hybride; EUR 90 000 - 140 000; Plein temps

Avantages offerts par ce poste

Equity

Hybrid work in Paris

Top-tier health insurance in France

L&D budget & hardware/tools

Team off-sites

Résumé du poste

White Circle, an AI Safety company, is hiring a ML Infrastructure Engineer to design and run distributed RL and post-training systems in production. The role focuses on scalable pipelines, data control for rollouts and policy updates, and end-to-end tuning for throughput and reliability.

You will build infrastructure for model iteration, evaluation, dashboards, and agent-centric development environments, with a hybrid Paris-based relocation.

Qualifications

• Experience designing and running distributed RL/post-training systems at scale.
• Strong Python and PyTorch or JAX.
• Debugging distributed GPU workloads across CUDA, drivers, containers, NCCL, networking, storage, and checkpointing.
• Profiling across the stack with py-spy, PyTorch profiler, Nsight, perf, tracing.
• Relocation to Paris (hybrid) required.

Responsabilités

• Build scalable RL and post-training pipelines, including smoke tuning runs for quality testing and ablations.
• Design data control systems for rollouts, replay, filtering, evaluation, and policy updates.
• Tune training and inference end-to-end for throughput: networking, memory, scheduling, data loading, storage, checkpointing.
• Build infrastructure for model iteration (experiment runs, artifacts, evals, dashboards, reproducibility, cost visibility) and inference infrastructure for post-training.
• Build agentic development environments: coding-agent harnesses, tool integrations, runtime sandboxes, multi-agent orchestration.

Connaissances

Python

PyTorch/JAX

Distributed RL

GPU debugging

Profiling tools

Performance optimization

NVIDIA CUDA

Outils

CUDA

NCCL

Slurm

Ray

Kubernetes

Nsight

Description du poste

White Circle, an AI Safety company, is hiring a ML Infrastructure Engineer to design and run distributed RL and post-training systems in production. The role focuses on scalable pipelines, data control for rollouts and policy updates, and end-to-end tuning for throughput and reliability.

You will build infrastructure for model iteration, evaluation, dashboards, and agent-centric development environments, with a hybrid Paris-based relocation.