Principal AI Researcher

📋 Description Lead design, training, and evaluation of LLMs for translation. Build evaluation frameworks beyond BLEU, TER, and MQM. Architect agentic NLP workflows: multi-step reasoning and tool use. Own the flagship model program from data curation to production. Design experiments with reproducible ML pipelines. Partner with Product/Engineering to ship capabilities. 🎯 Requirements Hands-on experience building/training LLMs or large transformers. Strong NLP background: MT, MT evaluation, IE, text generation, search, or RL. Design, implement, and evaluate agentic systems: tool-calling agents, multi-agent pipelines. Experience designing rigorous evaluation frameworks with automatic metrics and human-in-the-loop. Proficient in Python; libraries (Hugging Face, PyTorch, vLLM); workflow tools (Flyte/Prefect/Airflow). PhD in CS/Computational Linguistics or equivalent research output. 🎁 Benefits Global SaaS with international teams in Europe, APAC, Americas. Agile environment encouraging smart risks. Trust-based, supportive culture with open feedback and diversity. 4 company holidays + Christmas break. Your birthday off and 2 giveback days. Professional onboarding and Phrase learning.