Collective Communication Library Engineer (Software)

Rebellions is seeking a Collective Communication Library Engineer to design and implement key components of a new communication library tailored for their NPU's unique architecture. The role involves contributing to technical design, API definition, and performance optimization, as well as collaborating with hardware and software teams to analyze bottlenecks and influence future NPU architecture. Candidates should have a Master's degree and at least 5 years of experience in high-performance systems software development, with a strong understanding of collective communication algorithms, C/C++ programming, OS internals, networking fabrics, and interconnect architectures.

Must Have

• Master’s degree in Computer Science, Computer Engineering, or a related field
• Minimum of 5 years of professional experience in high-performance systems software development
• Strong collaboration and problem-solving skills for complex technical issues
• Solid understanding of collective communication algorithms (e.g., All-Reduce, All-Gather, Reduce-Scatter) and their performance characteristics
• Proficiency in low-level systems programming (C/C++) and understanding of OS internals and networking fabrics like RDMA/RoCE
• Understanding of interconnect topologies and Network-on-Chip (NoC) architectures
• Proven experience developing and delivering complex, high-performance, and reliable software in a collaborative environment

Good to Have

• A Ph.D. in a related field (HPC, Parallel Computing, Computer Architecture)
• Prior experience contributing to high-performance communication libraries (e.g., NCCL, MPI) or parallel runtimes
• Experience with performance analysis and optimization for AI accelerators (GPUs, TPUs, or other NPUs) and their specific interconnects (e.g., NVLink, CXL, RoCE)

Responsibilities and Opportunities

• Designing and implementing key components of a new collective communication library, specifically engineered for our NPU’s unique architecture and topology
• Contributing to the technical design, API definition, and performance optimization of the communication library
• Collaborating with hardware and software teams to analyze performance bottlenecks and influence future NPU and interconnect architecture

Key Qualifications

• Master’s degree in Computer Science, Computer Engineering, or a related field
• Minimum of 5 years of professional experience in high-performance systems software development
• Strong collaboration and problem-solving skills for complex technical issues
• Solid understanding of collective communication algorithms (e.g., All-Reduce, All-Gather, Reduce-Scatter) and their performance characteristics
• Proficiency in low-level systems programming (C/C++) and understanding of OS internals and networking fabrics like RDMA/RoCE
• Understanding of interconnect topologies and Network-on-Chip (NoC) architectures
• Proven experience developing and delivering complex, high-performance, and reliable software in a collaborative environment

Ideal Qualifications

• A Ph.D. in a related field (HPC, Parallel Computing, Computer Architecture)
• Prior experience contributing to high-performance communication libraries (e.g., NCCL, MPI) or parallel runtimes
• Experience with performance analysis and optimization for AI accelerators (GPUs, TPUs, or other NPUs) and their specific interconnects (e.g., NVLink, CXL, RoCE)

Application Process

• Document Screening > Online Interview (including coding test) > Culture-fit Interview > On-site Interview > Compensation Negotiation > Final Offer
• The application process may vary by job, and may change depending on schedule and circumstances.
• Application schedule and results will be individually notified via the email address provided during application.

Notes

• This announcement may close early when recruitment is complete.
• If there are false statements in the application, admission may be canceled.
• If the legal qualifications required for recruitment and job performance are not met, recruitment may be restricted.
• Veteran status and disability status do not disadvantage applicants in the recruitment process.
• The scope of duties may be changed considering the candidate's overall career and experience. If such changes are necessary, they will be communicated to the candidate at an appropriate time before the final offer.
• For recruitment inquiries, please contact the email address below.
• recruit@rebellions.ai