Cognition

Research Engineer, Infrastructure role at Cognition, an applied AI lab building end-to-end software agents like Devin. The role focuses on building and owning the core systems that researchers depend on, including distributed training infrastructure, experiment orchestration, data pipelines, and tooling to accelerate research velocity. This involves ensuring systems are fast, reliable, and scalable for large-scale training jobs across thousands of GPUs, with a focus on performance optimization and parallelism strategies. The ideal candidate has deep experience in distributed systems, Python/C++, PyTorch, GPU profiling, and ML knowledge to engage with researchers.