Principal ML Researcher - Remote, ESOP & PTO

Toloka; Madrid; Remote; EUR 120.000 - 160.000; Vollzeit

Benefits dieser Stelle

Remote or hybrid model

IT setup and home office allowances

Paid PTO and benefits vary by location

ESOP

Zusammenfassung

Toloka is seeking a Principal ML Researcher to define the technical vision and architecture for its core ML and post-training stack. You will bridge frontier AI research with platform engineering and lead efforts across RL, evaluation, and distillation at scale.

You will mentor senior engineers, collaborate with executives, and represent Toloka in the AI research community. A remote-friendly role with global impact.

Qualifikationen

• 6+ years in ML engineering or applied research, with 3+ years leading LLM post-training/fine-tuning at scale.
• Expertise in LLM alignment techniques: SFT, LoRA/PEFT, RLHF/RLAIF (GRPO, DPO, PPO).
• Strong evaluation/calibration experience and production-ready pipelines.

Aufgaben

• Define multi-quarter strategy and architecture for Toloka's post-training stack.
• Lead greenfield post-training & RL pipelines; scale RL and reward modeling.
• Design automated evaluation ecosystems with LLM-as-a-judge concepts.
• Advance model distillation, quantization, and context compression for latency/cost.
• Architect autonomous agents and tool-use workflows for multi-step reasoning.
• Mentor Senior ML Researchers and influence cross-functional roadmaps.
• Collaborate with Product/Engineering to translate client challenges into scalable architecture.

Kenntnisse

LLM alignment

SFT/LoRA/PEFT

RLHF/RLAIF

Python

PyTorch

Distributed training

LLM evaluation

Leadership

English communication

GPT/LLM deployment

Tools

Deepspeed

Megatron

FSDP

vLLM

TensorRT-LLM

Jobbeschreibung

Toloka is seeking a Principal ML Researcher to define the technical vision and architecture for its core ML and post-training stack. You will bridge frontier AI research with platform engineering and lead efforts across RL, evaluation, and distillation at scale.

You will mentor senior engineers, collaborate with executives, and represent Toloka in the AI research community. A remote-friendly role with global impact.