AI Frameworks Engineer

Start free →

Or apply yourself on the official page →

About the role

Job Details: Job Description: The Role and Impact Intel is building the open software stack that makes large language models run fast on Intel GPUs. You will join the engineering team that develops and optimizes LLM inference frameworks - working directly in vLLM and in the GPU kernel libraries beneath it, upstream and in the open. This is a deeply technical, inward-facing engineering role: your work is the framework itself. You will write and tune compute kernels, reshape execution paths, hunt performance bottlenecks down to the hardware, and land your changes in widely used open-source projects. If you want your code to decide how fast an entire class of AI workloads runs, this is that role. Business group The Data Center Group (DCG) is at the forefront of delivering cutting-edge technologies to power the world's data centers, enabling advanced computing solutions that drive innovation across industries. Within this group, our team is dedicated to developing AI frameworks and workloads that empower researchers and developers to create high-performance, scalable solutions for complex AI challenges. We aim to build tools and technologies that accelerate progress in AI, aligning with Intel's mission to advance global technology. Key Responsibilities - Design, develop, and optimize AI framework software for LLM inference on Intel GPUs, contributing upstream to external open-source frameworks such as vLLM and PyTorch. - Develop and optimize compute kernels (e.g., Triton, SYCL) for attention, mixture-of-experts, quantization, normalization, and other core building blocks of transformer inference. - Implement and tune distributed execution paths - tensor, pipeline, and data parallelism, collective communication, and asynchronous data movement within deep learning frameworks. - Transform and optimize the computational-graph representation of neural network models, and develop machine learning primitives in mathematical and kernel libraries. - Profile inference workloads to identify performance bottlenecks across the stack, quantify them, and propose and drive solutions together with the individual component teams that own each layer. - Optimize code for a range of computing hardware backends, adapting implementations to the memory hierarchy, instruction set, and execution model of each target. - Build and extend unit tests, correctness checks, and performance regression coverage so that framework and kernel work stays reliable as upstream moves. - Collaborate with framework, compiler, and machine learning engineers across global teams, and engage with the open-source community through code review and technical discussion. Qualifications: Minimum Qualifications - Bachelor's degree in AI, Computer Science, Computer Engineering, Mathematics, Physics, or a related field. - At least 2-3 years of relevant experience with a Bachelor's degree, or 1+ year with a Master's degree. - Advanced expertise in Python programming. - Demonstrated proficiency in PyTorch, including its internals or extension mechanisms. - Hands-on experience with AI inference and large language models (LLMs). - Strong problem-solving skills with experience in debugging and designing optimized software and algorithms. - Proficiency in Linux environments from a user perspective. - Solid foundation in AI frameworks, machine learning algorithms, and software engineering practices. Preferred Qualification - Experience writing or optimizing GPU compute kernels (Triton, CUDA, SYCL, OpenCL, or similar). - Experience in performance profiling and optimization for AI workloads, including reading and interpreting hardware-level profiles. - Familiarity with parallel and distributed computing concepts and collective communication libraries. - Familiarity with vLLM or another L

View the official posting →