Senior AI Research Engineer — Post-Training & Scale

Giotto.ai; Lausanne; Remote; CHF 180.000 - 240.000; Vollzeit

Zusammenfassung

Giotto.ai in Switzerland seeks a Senior Research Engineer or Research Scientist to own the training and optimisation of our post-training stack from pretrained checkpoints to production models. You will design, implement, scale, and operate supervised fine-tuning, RLHF, reward integration, and policy distillation.

Collaborating with data, evaluation, infrastructure, and inference teams, you will build scalable rollout systems, multi-machine training, and reproducible research software, while

Qualifikationen

• Proficient in Python, PyTorch and distributed execution for large-scale models.
• Experience with post-training methods like SFT, RLHF, and reward models is highly desirable.
• Ability to design scalable training pipelines across multiple machines and accelerators.

Aufgaben

• Own the 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, and process-based rewards.

Kenntnisse

Python

Distributed training

Experiment design

Research ownership

Problem solving

Tools

PyTorch

DeepSpeed

Megatron-Core

Hugging Face Transformers

CUDA

RAY

Jobbeschreibung

Giotto.ai in Switzerland seeks a Senior Research Engineer or Research Scientist to own the training and optimisation of our post-training stack from pretrained checkpoints to production models. You will design, implement, scale, and operate supervised fine-tuning, RLHF, reward integration, and policy distillation.

Collaborating with data, evaluation, infrastructure, and inference teams, you will build scalable rollout systems, multi-machine training, and reproducible research software, while