AI Software Development Engineer

About the role

Job Details:

Job Description:

The Role and Impact

We are seeking a skilled AI Software Development Engineer to join our team in optimizing AI inference workloads for Intel GPUs. In this role, you will work across the full AI software stack, focusing on improving performance-critical GPU kernels, graph compilation flows, and inference runtimes. Your contributions will directly enhance the efficiency and scalability of modern AI workloads, helping unlock the full potential of Intel's AI technologies.

Business Group

As part of Intel's Software Group, you will join a team dedicated to delivering innovative software solutions that enable and enhance Intel's AI products. The group plays a pivotal role in developing tools, frameworks, and optimizations that empower developers worldwide, shaping the future of AI-driven applications and advancing Intel's mission to create world-changing technology.

Key Responsibilities

- Optimize AI inference workloads, such as Large Language Models (LLMs), multimodal models, and emerging generative AI applications, for Intel GPUs.

- Design, develop, and tune performance-critical GPU kernels and runtime components using C++, SYCL, Vulkan Compute, OpenCL, or similar parallel programming technologies.

- Develop and optimize graph compilation flows and compiler passes using technologies like MLIR, LLVM, or related frameworks.

- Identify and resolve performance bottlenecks across compiler, runtime, framework, and kernel layers.

- Profile, benchmark, and analyze AI workloads to validate performance improvements and guide optimization strategies.

- Collaborate with hardware, driver, compiler, and framework teams to drive hardware/software co-optimization.

- Contribute to and optimize open-source AI inference frameworks such as OpenVINO, llama.cpp, vLLM, PyTorch, or related ecosystems.

Qualifications:

Minimum Qualifications

- Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field, with 4+ years of relevant industry experience, OR Master's degree with 3+ years, OR Ph.D. with no prior professional experience.

- Strong proficiency in C++ programming, debugging, and problem-solving.

- Solid understanding of GPU architectures, Graphics Device Driver, Firmware, AI accelerators, or parallel computing systems.

- Experience with performance analysis, profiling, and optimization for compute-intensive applications.

- Hands-on experience with modern neural network architectures and their execution on hardware accelerators.

Preferred Qualifications

- Experience optimizing end-to-end AI inference workloads in production environments.

- Familiarity with AI inference frameworks such as OpenVINO, PyTorch, TensorRT, ONNX Runtime, or similar technologies.

- Knowledge of neural network optimization techniques, including quantization, kernel fusion, and memory optimization.