Meta is looking for software engineers to help scale and improve the efficiency of large AI/ML workloads. A part of this is enabling high-performance interconnect (HPI) solutions, and optimizing bandwidth and latency to improve machine learning model performance from the top to the bottom of the stack.
This is an opportunity to work within a highly skilled team, collaborating with a large set of cross-functional partners and help bring next-generation large cluster architectures to life.
Responsibilities:
- Develop Linux kernel and firmware drivers for high-performance interconnect
- Supporting multi-platform environments, Linux and embedded
- Work on application support for RDMA-based communication
- Work closely with other teams to co-develop, drive features and innovations
- Analyze and improve efficiency, scalability, and stability with a system / end-to-end view
- Conduct code design and reviews
Minimum Qualifications:
- Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
- Deep understanding of Linux operating system internals, networking drivers, and development tools
- Experience working with hardware engineers on software support for custom designs
- Knowledge of PCIe, DMA controllers, networking hardware
- 3+ years of relevant coding experience in C, C++, or other relevant coding languages
- English language proficiency
Preferred Qualifications:
- Experience with high-performance interconnect standards, such as RoCE/Infiniband
- Experience developing or working with emulators/simulators, such as QEMU
- Knowledge of HPC workloads and systems at scale