Staff Research Engineer (Pre-training)

JetBrains is a company dedicated to creating robust and effective developer tools that automate routine tasks and empower creators. We are currently building an ambitious platform to integrate advanced AI capabilities across our entire product suite, and we are looking for a Staff Research Engineer to join our team in developing foundation models for coding tasks.

Key responsibilities

• Convert business requirements into technical specifications in collaboration with stakeholders.
• Train Large Language Models from scratch using large-scale GPU clusters.
• Collect, curate, and process high-quality pre-training and fine-tuning datasets.
• Deploy and maintain production-ready ML systems and subsystems.

Requirements

• Strong theoretical background in NLP and transformer-based architectures.
• Proven experience in the design, deployment, and support of production ML systems.
• Proficiency with modern deep learning frameworks such as PyTorch and common NLP libraries.
• Experience in distributed training of multi-billion parameter models.
• Excellent communication skills and a high level of attention to detail.

What we offer

• Access to a powerful training infrastructure featuring hundreds of NVIDIA GPUs.
• A modern ML stack including Python, PyTorch, HuggingFace, and Kubeflow.
• An inclusive environment that values diverse perspectives and independent problem-solving.
• The opportunity to work on cutting-edge AI technology that impacts developers globally.