AI Research Scientist

Overview

Join a world-class AI research lab to advance foundation models, native multimodal intelligence, and agentic systems. You will tackle core research areas and collaborate across teams to push frontier model capabilities. Expect work spanning model design, pre-training paradigms, data pipelines, evaluation, and inference optimization with real-world impact. This role offers opportunities to publish in top venues and contribute to cutting-edge AI research.

Responsibilities

• Conduct research on foundation models and native multimodal systems
• Develop and scale pre-training and fine-tuning methods (SFT, RLHF, DPO)
• Design data engineering pipelines and synthetic dataGeneration
• Build and evaluate agentic frameworks with reasoning and tool use
• Create robust evaluation benchmarks and safety/alignment metrics
• Optimize inference performance (quantization, distillation)
• Publish results in top conferences and contribute to open-source
• Collaborate with cross-functional teams to drive frontier model development

Key requirements

• PhD or Master’s in CS/AI/ML/Physics/Maths or equivalent research experience
• Strong hands-on expertise in at least one core area above
• Strong Python and PyTorch/distributed training experience
• Track record of top-tier publications (NeurIPS, ICML, ICLR, CVPR, ACL) or major open-source foundation model contributions

• collaboration
• clear communication
• problem-solving
• Python
• PyTorch/distributed training
• foundation model development

…