Software Engineer
Core
Developing the Collectives Communication Library to enable users to utilize large computing clusters for AI hardware.
Role type
Senior IC software engineer (systems/hardware integration)
Builds
High-bandwidth, low-latency communication primitives for AI hardware accelerators
Domain
AI compute infrastructure / Semiconductor software stack
Deliverable
production ML models | product features
Required skills
C++, Python, C, unit testing frameworks (Boost.Test, Google Test), build tools (CMake, Make, Ninja), version control (Git)
Preferred skills
RDMA networking libraries (libibverbs, libfabric), multithreading and parallel computing, inter-process communication (IPC), CI/CD pipelines, communication libraries (NCCL, MPI), machine learning frameworks (PyTorch), modern C++ standards (17/20)
Responsibilities
Implementing, testing and documenting the Collectives Communication Library; Collaborating with teams to design and test new features; Troubleshooting complex technical issues; Ensuring seamless integration of new hardware with the AI ecosystem
Seniority
Senior, hands-on IC
