Software Development Engineer II, EFA Network - Annapurna Labs
Core
Develop high-performance user-space software for the Elastic Fabric Adapter (EFA) network card to enable Machine Learning and HPC workloads on AWS.
Role type
Senior IC software engineer (high-performance networking)
Builds
Open source projects (Libfabric, Open MPI) and EFA user-space software stack
Domain
Cloud infrastructure, High-Performance Computing (HPC), Machine Learning infrastructure
Deliverable
production ML models
Required skills
C programming, high-performance software development, system design and architecture, open source contribution, API design, network software stack development, test development, stakeholder collaboration
Preferred skills
Full software development life cycle experience, network software stack optimization, cutting software stack occupancy
Technologies
C, Libfabric, Open MPI, EFA
Responsibilities
Write high-performance C code for open source EFA projects, invent new APIs for cloud networking concepts, provide expert-level support to major AI customers, design new ways to reduce software stack occupancy, write comprehensive tests for new features and regression prevention, collaborate with ML Infrastructure teams on cluster performance
Seniority
Senior, hands-on IC