Senior Research Engineer — Post-Training ML Systems (Remote)

Giotto.ai; Lausanne; Vor Ort; CHF 140.000 - 210.000; Vollzeit

Benefits dieser Stelle

Remote work supported

Swiss office meetups

Competitive compensation

Zusammenfassung

Giotto.ai in Switzerland is seeking a Senior Research Engineer or Research Scientist to own the training and optimisation side of our post-training stack. You will design, implement, scale, and operate methods to produce capable, reliable, and controllable production models.

The role covers supervised fine-tuning, preference optimisation, reinforcement learning, reward integration, policy distillation, and distributed training across multi-node hardware. PhD not required; strong ownership valued.

Qualifikationen

• Ownership of large-scale language-model training or post-training runs.
• Experience with workloads requiring multi-node distributed training.
• Strong proficiency in Python and PyTorch, including distributed execution.
• Experience with PyTorch Distributed, FSDP, and DeepSpeed or equivalents.
• Ability to select parallelism and sharding based on constraints.

Aufgaben

• Own end-to-end post-training pipeline from pretrained checkpoint to production candidate.
• Design and execute full-parameter and parameter-efficient SFT.
• Implement preference optimisation, RLHF, RLAIF, and related methods.
• Develop training strategies for reasoning, coding, tool use, multilingual behaviour, and long-horizon tasks.
• Integrate reward models, verifiers, critics, graders, and rewards.
• Build scalable rollout-generation systems for iterative training.
• Scale training across multiple machines and accelerators.
• Profile and improve accelerator utilisation, data loading, and training time.
• Diagnose instability, memory issues, and convergence problems.

Kenntnisse

Language-model training

Distributed systems

Python

PyTorch

Experiment design

Research ownership

Ausbildung

PhD not required

Tools

PyTorch Distributed

FSDP

DeepSpeed

Megatron-Core

Hugging Face Transformers

CUDA

NCCL

vLLM

Ray

Slurm

Kubernetes

MLflow

Weights & Biases

Docker

Jobbeschreibung

Giotto.ai in Switzerland is seeking a Senior Research Engineer or Research Scientist to own the training and optimisation side of our post-training stack. You will design, implement, scale, and operate methods to produce capable, reliable, and controllable production models.

The role covers supervised fine-tuning, preference optimisation, reinforcement learning, reward integration, policy distillation, and distributed training across multi-node hardware. PhD not required; strong ownership valued.