Overview

MLIR in Action: A Practical Guide to Scalable Model Optimization and Hardware Acceleration with OpenXLA, IREE, and Mojo Unlock the full power of machine learning optimization and next-generation compiler design with MLIR in Action - your complete, hands-on guide to mastering the Multi-Level Intermediate Representation (MLIR) ecosystem. Built for engineers, researchers, and AI practitioners, this book takes you on a step-by-step journey through the core concepts, workflows, and real-world applications of MLIR - the backbone of modern compiler infrastructures like OpenXLA, IREE, and Mojo. Learn how to optimize, transform, and deploy machine learning models efficiently across CPUs, GPUs, and custom accelerators using a unified and extensible compiler stack. Inside this practical guide, you will discover how to: - Understand the architecture and principles of MLIR in depth. - Build, extend, and debug custom MLIR dialects and passes. - Integrate MLIR with leading frameworks such as TensorFlow, PyTorch, and Mojo. - Leverage OpenXLA and IREE for portable, high-performance model deployment. -Automate builds, CI/CD pipelines, and cloud deployment for scalable production systems. -Visualize, profile, and debug IR flows for performance tuning and optimization. - Stay ahead of the curve with insights into emerging compiler standards, AI-driven optimization, and future MLIR trends. From theory to practice, every chapter blends clear explanations with real code examples, text-based flowcharts, and implementation checklists. Whether you are optimizing large-scale AI workloads or exploring compiler-based acceleration, MLIR in Action gives you the tools to move from concept to production with confidence. Perfect for: - Machine Learning Engineers - Compiler Developers - AI Infrastructure Architects - Systems Programmers - Researchers exploring hardware-aware AI optimization MLIR in Action bridges the gap between research and real-world deployment - helping you build the scalable, efficient, and future-ready AI systems that power the next wave of machine learning innovation.

Full Product Details

9798272080059

Table of Contents

Reviews

Author Information

Tab Content 6

Countries Available