Lead Research Engineer: Remote Post-Training & RL AI

Giotto.ai SA; Lausanne; Hybrid; CHF 140.000 - 190.000; Vollzeit

Zusammenfassung

Giotto.ai SA, a Switzerland-based AI company, seeks a Lead Research Engineer to own the training and optimization stack for production models. You will design, implement, and operate post-training methods across multiple machines, focusing on SFT, RLHF, and reward-based strategies.

Remote work is supported with in-person team meetings in Europe. You will guide researchers and engineers, set technical direction, and ensure robust, scalable, and reproducible training pipelines across distributed

Qualifikationen

• Ownership of large-scale language-model training or post-training runs across multiple machines and accelerators.
• Experience with workloads where single-node training is insufficient.
• Strong proficiency with Python and PyTorch.
• Experience with PyTorch Distributed, FSDP, DeepSpeed, or Megatron-Core.

Aufgaben

• Own the end-to-end post-training pipeline from pretrained checkpoint to production candidate.
• Set technical direction and priorities for post-training and RL work.
• 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 agent tasks.
• Integrate reward models, verifiers, critics, graders, and rewards.
• Build scalable rollout-generation systems for iterative and on-policy training.
• Scale training across multiple machines and accelerators.

Kenntnisse

Python

PyTorch

Distributed execution

SFT

Reinforcement learning

RLHF

Kubernetes

Megatron-Core

DeepSpeed

CUDA

NCCL

Hugging Face Transformers

vLLM

Rollout engines

FSDP

Ray

Slurm

Docker

Tools

PyTorch Distributed

FSDP

DeepSpeed

Megatron-Core

CUDA

NCCL

Jobbeschreibung

Giotto.ai SA, a Switzerland-based AI company, seeks a Lead Research Engineer to own the training and optimization stack for production models. You will design, implement, and operate post-training methods across multiple machines, focusing on SFT, RLHF, and reward-based strategies.

Remote work is supported with in-person team meetings in Europe. You will guide researchers and engineers, set technical direction, and ensure robust, scalable, and reproducible training pipelines across distributed