Senior Research Scientist | Model Scaling

Job Summary

Required Qualifications

• •Strong hands-on experience adapting and scaling large language models via fine-tuning, instruction-tuning, or post-training of multi-billion-parameter models beyond black-box use
• •Sound judgment about architecture trade-offs at scale (e.g. dense vs. MoE) and about which open-weight foundation models to build on
• •Working knowledge of parameter-efficient and multi-capability adaptation (LoRA/PEFT and variants)
• •A hands-on builder who enjoys training models, running experiments, and debugging pipelines, and who can carry research results through to production with engineering
• •Strong coding and experimentation skills (Python, PyTorch/JAX/Tensorflow)
• •Ability to communicate clearly, collaborate across teams, and align research work with product and engineering priorities
• •Drive the selection and evaluation of open foundation / open-weight models as the basis for our next-generation translation systems
• •Lead model selection and general architecture decisions for scaling to hundreds of billions of parameters, including Mixture-of-Experts and other sparse or efficient designs
• •Design multi-capability adaptation strategies using LoRA, PEFT, and related methods
• •Own the modelling lifecycle for your work: prototyping, ablations, scaling experiments, evaluation, and delivery into production, with rigorous and reproducible evaluation
• •Partner closely with post-training, RL/RLHF, and instruction-following specialists to integrate alignment and capability work into the base model
• •Stay ahead of the open-model and scaling literature, and bring well-founded recommendations back to the team

Desired Qualifications

• •Experience quantifying uncertainty in large models — calibration and confidence estimation via Bayesian methods, ensembling, steering, or prompt-based approaches
• •Experience with machine translation, multilingual NLP, or document-/layout-aware modelling experience
• •Familiarity with MoE-specific training and adaptation (e.g. expert routing, Mixture-of-LoRA-Experts) and large-scale data-mixture design