AI back-end Engineer

Introduction

At IBM Infrastructure & Technology, we design and run the systems that keep the world operating. Our work spans high-resiliency mainframes, hybrid cloud platforms, networking, automation, and site reliability. We deliver the performance, security, and scalability that clients and industries rely on daily. In Infrastructure & Technology, you tackle sophisticated challenges through curiosity and collaboration, working with a wide range of technologies and colleagues worldwide to build resilient, future-ready solutions that enable innovation. With continuous learning, career advancement, and a collaborative culture, IBM supports you in developing expertise and shaping the infrastructure that powers progress.

Your Role And Responsibilities

As a Back-End Developer focused on Artificial Intelligence for IBM Z, you will collaborate in an agile, team-oriented environment to design, code, and deliver innovative solutions to client needs. You will use IBM's Design Thinking to build solutions that combine excellent user experience with world-class quality, resiliency, performance, security, and stability. Your core responsibilities include:

• Design and Develop Solutions: Design, implement, debug, test, and deliver creative solutions for client problems and requirements across a variety of AI-related areas, using DevOps methodologies.
• Collaborate with Teams: Participate in an agile, collaborative setting to understand requirements and deliver high-quality solutions aligned with client needs.
• Apply Design Thinking: Use IBM's Design Thinking to create solutions that deliver an exceptional user experience plus world-class quality, resiliency, performance, security, and stability.
• Lead Edge Projects: Contribute to leading-edge initiatives using technologies such as machine learning, deep learning, and other AI-related areas to drive innovation and accelerate solution delivery.
• Ensure Solution Quality: Confirm solutions meet required standards for quality, resiliency, performance, security, and stability.
  
  

Required Technical And Professional Expertise

• 5-10 years of industrial experience developing and working with Unix/Linux based products using Python/C/C++.
• AI Development Experience: Experience designing, coding, debugging, and testing PyTorch Core & Internals.
• Contribution to and ability to extend PyTorch internals and backend architecture, including:
• TorchInductor
• TorchDynamo
• AOTAutograd
• Debug low-level PyTorch runtime issues (autograd, dispatcher, kernels, graph breaks).
• Deep hands-on experience with PyTorch FSDP
• Optimize model execution paths for both training and inference.
• Open-source contributions to PyTorch or related projects
• Experience with proprietary or custom AI accelerators
• Experience integrating AI frameworks with custom or proprietary accelerators
• Advanced skills in C++ and Python
• Experience with GPU programming or accelerator SDKs
• Advanced Python skills and strong debugging ability
• DevOps Methodologies: Experience applying DevOps methodologies to deliver high-quality solutions on AI frameworks while ensuring world-class quality, resiliency, performance, security, and stability.
  
  

Preferred Technical And Professional Experience

• Deep understanding of:
• AI runtimes
• Memory hierarchies
• Parallel execution models
• PyTorch distributed runtime
• Parameter sharding and memory management
• Hands-on experience with torch.compile and TorchInductor
• Design Thinking Application: Experience applying IBM's Design Thinking to create solutions with a great user experience, with an emphasis on delivering innovative and effective solutions.
• Technical Leadership: Experience leading or contributing to leading-edge projects, driving innovation and solution delivery in Core backend AI areas.
• Excellent communication skills