ML Infrastructure Engineer

Responsibilities

We are representing a confidential AI safety organization in Paris that is seeking an ML Infrastructure Engineer to join their technical team. In this role, you will:

• Build scalable reinforcement learning 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 optimization across networking, memory, scheduling, data loading, storage, checkpointing, and I/O
• Build infrastructure supporting model iteration including experiment runs, artifacts, evaluations, dashboards, reproducibility tracking, and cost visibility
• Develop inference infrastructure for post-training and evaluation loops
• Build agentic development environments with coding-agent harnesses, tool integrations, runtime sandboxes, and multi-agent orchestration

What We're Looking For

Required qualifications:

• Hands-on experience designing and running distributed RL/post-training systems at scale, including rollouts, replay buffers, reward signals, policy updates, and evaluation loops
• Strong Python skills including concurrency, async, multiprocessing, and performance optimization
• Deep expertise in PyTorch or JAX
• Proven ability debugging distributed GPU workloads across CUDA, drivers, containers, NCCL, networking, storage, and checkpointing
• Profiling experience across the stack using tools such as py-spy, PyTorch profiler, Nsight, perf, and tracing frameworks
• Experience with inference stacks such as vLLM, SGLang, TensorRT-LLM, Dynamo, or custom serving solutions
• Ability to connect system metrics to model behavior and learning dynamics
• Willingness to relocate to Paris for hybrid work

Bonus qualifications:

• Public builder footprint with open-source contributions to RL, distributed ML, inference, evaluation, or agent infrastructure
• Experience at high-caliber AI infrastructure or research teams
• Ownership of custom training frameworks, trainers, schedulers, or data loaders
• GPU cluster experience with Kubernetes, Slurm, or Ray; familiarity with NCCL, RDMA, InfiniBand, RoCE, or EFA
• Proficiency in Rust, C++, CUDA, or Go
• Serious use of agentic coding tools

How We Work With You

Our talent team at Get A Job.ai will guide you through the entire process. After you apply through our platform, one of our recruiters will conduct an initial screening to understand your background and ensure alignment with the role. We will then submit qualified candidates directly to our client. Please do not attempt to contact the client directly, as all communication should flow through Get A Job.ai to ensure a smooth and professional process.

Pay

Compensation details will be discussed during the screening process with our recruiting team.

The client offers competitive salary plus equity, a relocation package to Paris, hybrid work arrangements, top-tier medical insurance in France, flexible time off, learning and development budget, all necessary hardware and tools, covered AI agent and IDE subscriptions, and team off-sites twice per year.

Equal Opportunity

Get A Job.ai is committed to fair and equitable recruiting practices. We welcome applications from candidates of all backgrounds and work to ensure an inclusive process for everyone we represent.