Software Engineer, Collective Communication
Core
Design and implement custom networking collectives in C++ and CUDA to enable efficient training of large-scale AI models on custom supercomputers.
Role type
Senior IC software engineer (low-level networking)
Builds
High-performance collective communication stacks for AI training
Domain
AI infrastructure / High-performance computing
Deliverable
production ML models
Required skills
C++, CUDA, low-level performance critical software, distributed algorithms, RDMA, network simulation
Preferred skills
collective communication, supercomputer architecture
Technologies
C++, CUDA, RDMA
Responsibilities
Collaborate with ML researchers to design efficient collective operations; optimize network transport usage for large training jobs; develop network simulations for future supercomputer designs
Seniority
Senior, hands-on IC
Sourced via ashby · Listed on CareerPlan, which tracks 70,000+ jobs from 20+ sources.
