Senior Systems Software Engineer, LPU (Toronto)

NVIDIA’s LPU System Software team builds foundational software that enables deterministic, high-performance computing platforms by shifting complexity from silicon into software. We design and maintain the hardware abstraction layers, core system libraries, and runtime components that allow compiler teams and data center operators to safely and efficiently execute workloads on novel architectures. In this role, you will develop and evolve the libraries, drivers, and runtime interfaces that serve as key entry points into the platform. What You’ll Be Doing

Extend and maintain hardware abstraction layers and core system libraries used across the platform. Design and implement drivers, runtimes, and data movements/aggregation pipelines supporting workload execution. Build and maintain runtime interfaces for launching, monitoring, and managing workloads. Improve platform reliability through automation, error reporting, diagnostics, and operational tooling. Debug and resolve complex...