Senior MLOps Engineer: Scale LLM Training & Inference

multiversecomputing; Donostia/San Sebastián; Híbrido; EUR 70.000 - 110.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Indefinite contract

Equal pay guaranteed

Variable performance bonus

Signing bonus

Work visa sponsorship

Relocation package

Private health insurance

Hybrid opportunity

Flexible working hours

Language classes

Discounted meals

Descripción de la vacante

Multiverse Computing is seeking a Senior MLOps Engineer to shape the Training and Inference Optimization team. You will architect infrastructure powering next-gen AI models, bridging systems programming and ML to optimize large-scale training and ultra-high-throughput serving.

You will lead training pipelines, inference orchestration, and lifecycle management while mentoring engineers and driving cost-efficient, production-ready solutions.

Formación

• 5+ years in MLOps, DevOps, or Software Engineering, with at least 2 years in LLM infrastructure.
• Expert-level proficiency with PyTorch and the NVIDIA stack (CUDA, NCCL, Triton).
• Hands-on experience with NVIDIA NeMo (or Megatron‑Bridge) for distributed training and at least two serving options: vLLM, TensorRT-LLM, or SGLang.
• Experience with SLURM/Flyte/Ray/SkyPilot for cluster management and MLflow for experiment and model management.
• Deep expertise in Kubernetes and K8s operators (KubeRay, MPI Operator, Run:ai).
• Mastery of Python and knowledge of C++ or Rust for performance-critical components.
• Familiarity with high-performance networking (InfiniBand/RoCE) and NVIDIA H200/B200 (Blackwell).

Responsabilidades

• Training Infrastructure: Architect and maintain scalable distributed training pipelines using NVIDIA NeMo/Nemotron/Megatron‑Bridge.
• optimise GPU utilisation, manage checkpointing strategies, and implement automated fault tolerance for long-running jobs.
• Inference Orchestration: Lead the deployment of LLMs using vLLM, TensorRT‑LLM, or SGLang; tune PagedAttention, batching, and quantisation.
• Workload Orchestration: Use SLURM/Flyte/Ray/SkyPilot to manage ML workloads across clouds and on-prem clusters.
• Lifecycle Management: Standardise model tracking and versioning with MLflow.
• Performance Engineering: Profile and optimize across CUDA kernels, NCCL, and Python orchestration.
• Efficiency & Cost Governance: Optimize cloud/on-prem GPU expenditures with smart scaling.
• Technical Leadership: Drive roadmap, code reviews, and mentor teams.

Conocimientos

MLOps

DevOps

Software Engineering

LLM infrastructure

PyTorch

CUDA

NCCL

Triton

NVIDIA NeMo

Megatron-Bridge

vLLM

TensorRT-LLM

SGLang

SLURM

Flyte

Ray

SkyPilot

MLflow

Kubernetes

KubeRay

MPI Operator

Run:ai

Python

C++

Rust

InfiniBand

RoCE

NVIDIA H200

NVIDIA B200

Open-source contributions

Model compression

Triton kernels

Herramientas

NVIDIA NeMo

Megatron‑Bridge

vLLM

TensorRT-LLM

SGLang

Descripción del empleo

Multiverse Computing is seeking a Senior MLOps Engineer to shape the Training and Inference Optimization team. You will architect infrastructure powering next-gen AI models, bridging systems programming and ML to optimize large-scale training and ultra-high-throughput serving.

You will lead training pipelines, inference orchestration, and lifecycle management while mentoring engineers and driving cost-efficient, production-ready solutions.