Senior LLM Engineer: Multimodal AI, Compression & RL on GPUs

Hyperproof; Barcelona; Presencial; EUR 150.000 - 210.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Indefinite contract

Equal pay guaranteed

Variable bonus

Signing bonus

Visa sponsorship

Relocation package

Private health insurance

Educational budget

Hybrid opportunity

Flexible hours

Language classes

Discounted lunch

Career plan

Happy people culture

Descripción de la vacante

Multiverse Computing is seeking a Senior LLM Engineer to join the Research team in Barcelona. You will manage the full model lifecycle—from data preparation and distributed training to post-training, compression and efficient inference.

Priorities shift in research, so robust fundamentals and adaptability are valued over specialization. You will prototype post-transformer architectures, build data pipelines, and publish papers or code when appropriate.

Formación

• Master’s or PhD in Computer Science, AI, Mathematics, Physics or related field.
• Senior: 4+ years of hands-on experience training or post-training transformer models.
• Strong PyTorch. Comfortable working below library level: model code, custom losses, parallelism config.
• Distributed training (FSDP, DeepSpeed or Megatron-LM; data, tensor, pipeline and expert parallelism) on multi-node clusters with Slurm/K8s.
• Hands-on post-training experience: SFT, DPO, and at least one RL method (PPO, GRPO or similar).
• Solid mathematical foundations in linear algebra, optimisation and probability, and an understanding of training dynamics.
• Good understanding of GPU fundamentals: memory hierarchy, mixed precision, profiling, throughput.
• Rigorous experimental practice: ablation design, reproducibility, experiment tracking, noise-aware evaluation.
• Strong debugging and communication skills. Fluent English.

Responsabilidades

• Train and post-train LLMs and multimodal models on multi-node GPU clusters using SFT, preference optimisation, RL (e.g. GRPO), and on-policy and offline distillation.
• Design compression techniques (quantization, pruning, quantization-aware training, quantum-inspired tensor-network methods) and healing recipes that recover the quality of compressed models.
• Build RL environments, reward functions and verifiers for reasoning, coding and agentic tasks, and extend RL methods to world models and Physical AI.
• Research and prototype post-transformer and hybrid architectures (state-space models, linear attention, MoE), from small-scale ablations to scaling recipes.
• Build data pipelines for curation, filtering, deduplication and large-scale synthetic data generation.
• Design rigorous ablations and evaluations on current benchmark suites (reasoning, coding, long-context, instruction following, agentic) as well as internal evals.
• Port and optimise models for inference on various accelerators and edge chips, profiling latency, memory and throughput.
• Write up experiments and results as internal reports and, where relevant, papers and open-source releases.
• Review code, mentor teammates, and follow and bring in the latest research.

Conocimientos

Masters/PhD in CS/AI/Math/Physics

4+ years transformer training

Strong PyTorch

Distributed training

SFT/DPO/RL

Math foundations

GPU fundamentals

Experimental rigor

Fluent English

Educación

Master’s or PhD in CS/AI/Math/Physics

Herramientas

Slurm

Kubernetes (K8s)

Megatron-LM

DeepSpeed

NVIDIA Megatron

PyTorch

FSDP

Descripción del empleo

Multiverse Computing is seeking a Senior LLM Engineer to join the Research team in Barcelona. You will manage the full model lifecycle—from data preparation and distributed training to post-training, compression and efficient inference.

Priorities shift in research, so robust fundamentals and adaptability are valued over specialization. You will prototype post-transformer architectures, build data pipelines, and publish papers or code when appropriate.