Manjunath Kudlur
Papers
1
Total Citations
9,777
H-Index
1
About
Manjunath Kudlur is a pioneering figure in the development of large-scale machine learning infrastructure, best known for his foundational contributions to the TensorFlow ecosystem. As a key architect of TensorFlow, Kudlur co-authored the landmark 2016 paper "TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems," which has amassed over 9,700 citations. This work introduced a flexible, open-source framework that revolutionized how machine learning algorithms are expressed and executed across heterogeneous systems—from mobile devices to massive distributed clusters. Kudlur’s research focuses on compiler design, parallel computing, and domain-specific languages, enabling efficient deployment of deep learning models at scale. His work on TensorFlow’s core runtime and graph execution engine directly shaped the platform’s ability to handle complex computations with minimal code changes, making it a cornerstone of modern AI research and industry practice. Beyond TensorFlow, Kudlur has contributed to GPU programming models and high-performance computing, earning recognition as a leader in bridging the gap between hardware and software for machine learning. His innovations continue to empower researchers and engineers worldwide to build and deploy intelligent systems with unprecedented ease and efficiency.
Research Focus
Key Achievements
Top Papers
- 1TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems9,777 citations · 2016