Senior AI Software Development Engineer

Intel•Hillsboro, OR

1d•Hybrid

About The Position

Join Intel's Communications Runtime team to build the communication software stack powering large AI and High-Performance Computing (HPC) workloads. The role involves architecting, developing, and optimizing the oneAPI Collective Communications Library (oneCCL) to enable scalable, high-performance communication across Intel CPU and GPU platforms. Work on distributed systems challenges involving collective communication algorithms, communication runtimes, networking technologies, operating systems, and large-scale AI training infrastructure. Collaborate across software and hardware teams to deliver performance for AI model training, scientific computing, and distributed computing workloads. The role offers opportunities to influence Intel's software roadmap, contribute to open-source communities, and develop innovations that scale across processors and accelerators. The position is based in Hillsboro, Oregon, with an additional location in Austin, Texas, and is eligible for Intel's hybrid work model, splitting time between an assigned Intel site and off-site. The role is a Shift 1 position. Compensation includes competitive pay, stock bonuses, and benefit programs; the stated annual US salary range is $133,410–$240,710.

Requirements

Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field, with 3+ years of relevant experience.
Strong software development experience using C/C++.
Experience developing systems software, middleware, runtime software, networking software, or communication libraries.
Hands-on experience diagnosing and optimizing performance, scalability, and reliability in complex distributed systems.
Experience in distributed systems, operating systems, concurrency, networking, and computer architecture.
Experience developing software in Linux environments.

Nice To Haves

Experience with collective communication frameworks such as oneCCL, MPI, NCCL, RCCL, UCX, OpenSHMEM, or libfabric.
Experience supporting distributed AI training, machine learning infrastructure, or HPC applications.
Experience optimizing communication performance across multiple nodes, GPUs, or accelerators.
Familiarity with RDMA technologies and high-performance networking environments.
Experience developing software for GPU, accelerator, or heterogeneous computing platforms.
Open-source software development experience.
Customer-facing experience troubleshooting complex technical and performance-related challenges.
Experience with technologies including oneAPI, SYCL, PyTorch Distributed, DeepSpeed, Horovod, RDMA, InfiniBand, RoCE, Intel VTune Profiler, and GDB.

Responsibilities

Design, develop, and optimize collective communication algorithms and runtime capabilities within oneCCL.
Improve scalability, latency, bandwidth utilization, resilience, and communication performance across distributed systems.
Lead performance optimization initiatives for multi-node and multi-accelerator AI and HPC workloads.
Investigate and resolve complex system-level performance, reliability, and interoperability issues.
Collaborate with hardware architects, software engineers, researchers, open-source contributors, and strategic customers.
Contribute to and influence open-source communication and networking technologies.
Drive engineering decisions through data-driven analysis and performance insights.