ML Infra Engineer for Scalable RL & Post-Training

Visa Hunt; Paris; Hybride; EUR 85 000 - 135 000; Plein temps

Avantages offerts par ce poste

Relocation package

Comprehensive medical insurance (Paris

Hardware and tools provided

Subscriptions for AI agents/IDEs

Team off-sites twice a year

Résumé du poste

White Circle is seeking an ML Infrastructure Engineer to build the systems behind our LLM post-training, RL, evaluation, inference, and agentic development workflows. You’ll work close to researchers, GPUs, training loops, data control systems, evals, inference stacks, and the infrastructure decisions that directly affect model learning and product quality.

You’ll design scalable pipelines, tune end-to-end performance, and build agentic environments while staying close to state-of-the-art in

Qualifications

• Experience designing distributed RL/post-training systems at scale.
• Fluent in PyTorch or JAX.
• Proficient in Python with concurrency and performance optimization.

Responsabilités

• Build robust RL and post-training pipelines for quality testing and ablations.
• Design data control systems for model input, data flow, and policy updates.
• Tune training and inference end-to-end for high throughput on GPUs.
• Investigate how infra choices affect learning and eval quality.
• Build infrastructure for experiment runs, artifacts, evals, dashboards, and cost visibility.
• Develop agentic development environments and multi-agent orchestration.
• Collaborate with the team to plan steps and share context early.

Connaissances

Distributed RL systems

Python programming

Performance optimization

Ownership mindset

Debugging distributed GPUs

System metrics reasoning

Outils

PyTorch

JAX

Kubernetes

Slurm

Ray

TensorRT-LLM

vLLM

NCCL

UCX

CUDA

Go

Rust

Description du poste

White Circle is seeking an ML Infrastructure Engineer to build the systems behind our LLM post-training, RL, evaluation, inference, and agentic development workflows. You’ll work close to researchers, GPUs, training loops, data control systems, evals, inference stacks, and the infrastructure decisions that directly affect model learning and product quality.

You’ll design scalable pipelines, tune end-to-end performance, and build agentic environments while staying close to state-of-the-art in