AI Frameworks Engineer

Intel•Austin, CA

1d•Hybrid

About The Position

Join Intel's AI Frameworks PyTorch team to shape the future of Artificial Intelligence (AI) and High-Performance Computing (HPC). As an AI Frameworks Engineer, you will design, build, and optimize AI software frameworks that enable breakthrough scientific discoveries and machine learning innovations at scale. You will contribute to the success of the new Argonne AI Center of Excellence, a partnership between Intel and Argonne National Labs. This role shall focus on identifying performance bottlenecks and functional limitations in PyTorch, vLLM, and related AI framework software when running key AI workloads and models at large scale. After identifying issues that impact functionality, scalability, and performance, you will help design and implement optimizations that maximize the utility of Intel CPUs, GPUs, and AI accelerators. You will work as part of a larger team developing and optimizing next-generation AI framework software, enabling efficient training, inference, and deployment of modern AI models. Your innovative work will improve performance across Intel architectures and empower HPC and AI applications to achieve higher throughput, lower latency, and stronger developer productivity.

Requirements

Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Mathematics, or STEM-related field with 3+ yrs. of experience in software development.
OR Master's degree in Computer Science, Computer Engineering, Electrical Engineering, Mathematics, or STEM-related field with 1+ yrs. of experience in software development.
OR Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, Mathematics, or STEM-related field with 3+ months of experience in software development.
3+ years of experience in at least one of the following: AI framework development or optimization.
PyTorch, vLLM, TensorFlow, Hugging Face, DeepSpeed, or related AI software frameworks.
AI model development, enablement, profiling, or performance optimization.
GPU or accelerator software development.
Runtime, compiler, kernel, or backend optimization for AI workloads.

Nice To Haves

Advanced degree, Master's or PhD, in Computer Science, Computer Engineering, Electrical Engineering, Mathematics, or STEM-related field.
Proficiency in Python, SYCL/CUDA, and C++ programming.
Experience developing in Linux environments.
Background in AI framework internals, model execution, or backend integration.
Experience with PyTorch, vLLM, Hugging Face Transformers, DeepSpeed, or similar AI frameworks.
Experience optimizing AI training or inference workloads for CPUs, GPUs, or accelerators.
Background in performance profiling, memory optimization, throughput improvement, or latency reduction.
Experience enabling or optimizing large language models, generative AI models, or scientific AI models.
Understanding of deep learning algorithms, model architectures, and AI workload patterns.
Strong analytical skills and ability to solve complex software challenges.
Passion for driving meaningful advancements in AI software and scientific computing.

Responsibilities

Identify performance bottlenecks and additional features necessary to run Argonne AI CoE workloads.
Optimize PyTorch, vLLM, and related AI framework software for Intel CPUs, GPUs, and AI accelerators.
Enable and optimize key AI models, including large language models, generative AI models, and scientific AI workloads.
Profile AI training and inference workloads to identify issues across framework, runtime, kernel, and hardware layers.
Collaborate with cross-functional teams to define technical specifications and software requirements.
Troubleshoot and resolve complex issues across multiple hardware and software stack layers.
Contribute to software innovations that enhance AI framework performance, scalability, and usability.
Partner with engineering and architecture teams to maximize AI model performance on Intel architectures.