Research Engineer, Discovery

As a Research Engineer on the Discovery team, you will work end-to-end across the model stack to remove infra blockers on the path to scientific AGI.
You will design and implement large-scale infrastructure systems to support AI scientist training, evaluation, and deployment in distributed environments.
Identify bottlenecks, develop robust evaluation frameworks, and build scalable VM/sandboxing/container architectures to safely run long-horizon AI tasks.
Translate experimental requirements into production-ready infrastructure and develop large-scale data pipelines for advanced language model training.
Optimize training and inference pipelines for stable, efficient reinforcement learning and collaborate with researchers to scale ideas from experimentation to production.
Ideal candidates have 6+ years in infrastructure engineering, deep knowledge of high-throughput ML architectures, and hands-on experience across the ML stack.