Principal AI Researcher

📋 Description

• Lead design, training, and evaluation of LLMs for translation and quality (LoRA/DPO).
• Design evaluation frameworks for translation quality beyond BLEU, TER, MQM, and COMET.
• Architect agentic NLP workflows: multi-step reasoning, tool use, cross-model orchestration.
• Take technical ownership of flagship model development from data curation to production.
• Design and run ML experiments with reproducibility and clear results.
• Act as technical lead for the research team; mentor researchers; shape direction and hiring.

🎯 Requirements

• Deep hands-on experience building and training LLMs; pre-training, fine-tuning, RLHF/DPO.
• Strong NLP background with MT, MT evaluation, information extraction, text generation, or RL.
• Proven ability to design, implement, and evaluate agentic systems: tool-calling agents, multi-agent pipelines, or LLM-orchestrated workflows at scale.
• Experience designing and running rigorous evaluation frameworks, including automatic metrics and human-in-the-loop eval.
• Strong Python skills and ML/NLP libraries (Hugging Face, PyTorch, vLLM); familiarity with Flyte/Prefect/Airflow.
• PhD in CS or Computational Linguistics or equivalent research output.

🎁 Benefits

• Work in a growing global SaaS company.
• International team across Europe, APAC, Americas.
• Expert colleagues building the localization platform.
• Agile environment with smart risk-taking.
• Culture of trust, support, and openness.
• Professional development and learning opportunities.