Member of Technical Staff, Model Training (infra & platform)

In short

This full-time remote role is about building the systems that help AI researchers train bigger models and run large experiments. You would own key platforms, from GPU infrastructure to serving and simulation tools. It may suit experienced engineers who enjoy hard performance problems and making design choices that others rely on.

What you will do

• Design and build shared infrastructure for fine-tuning and reinforcement learning models.
• Create training pipelines that coordinate many GPU machines, data generation and simulations.
• Find slow spots in the system and make experiments run faster and use GPUs better.
• Set technical standards and lead big engineering projects from design to production, while mentoring other engineers.

What you need to apply

• Solid computer science basics, especially in distributed systems, networking and performance.
• At least 3 years of experience building scalable backend systems or distributed computing platforms.
• Good Python skills, plus a systems language such as C++, Rust or Go.
• Hands-on work with GPU machine learning infrastructure, such as distributed training or model inference.

Who can apply

The role is open worldwide, with no country restriction stated, and it is full time and remote. You need at least 3 years of relevant experience.