logo

For the Inventor, By the Inventor

Saratoga, CA, United States of America

Jindrich Zejda

This inventor holds 32 USPTO granted patents and 5 EPO patents, primarily in Neural Network Acceleration. Top assignees: Amazon Technologies, Inc., Xilinx, Inc.. Active years: 2017-2025.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Jindrich Zejda. Retrieved from https://idiyas.com/inventor/jindrich-zejda

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

USPTO Granted Patents = 32

% Patents Active = 100.0


Average Co-Inventor Count = 5.0

ph-index = 7

Forward Citations = 166(Granted Patents)


Company Filing History:

goldMedal17 out of 23,655 
 
Amazon Technologies, Inc.
 patents
silverMedal15 out of 5,108 
 
Xilinx, Inc.
 patents

Years Active: 2017-2025

Loading Chart...
Loading Chart...
Areas of Expertise:
Neural Network
Load Cloning
Operator Fusion
Memory Allocation
Parallel Execution
Debugging
Matrix Multiplication
Resource Allocation
Hardware Acceleration
Data Exchange
Saratoga, CA
Design Optimization
32 patents (USPTO):Explore Patents
1. 12400106 - Arithmetic-intensity based load cloning (2025-08-26),
2. 12321849 - Performing hardware operator fusion (2025-06-03),
3. 12182688 - Hierarchical partitioning of operators (2024-12-31),
4. 12093806 - Static memory allocation for neural network inference (2024-09-17),
5. 12086572 - Software defined neural network layer pipelining (2024-09-10),
6. 12061990 - Static block scheduling in massively parallel software defined hardware systems (2024-08-13),
7. 11809981 - Performing hardware operator fusion (2023-11-07),
8. 11694066 - Machine learning runtime library for neural network acceleration (2023-07-04),
9. 11620490 - Multi-layer neural network processing by a neural network accelerator using host communicated merged weights and a package of per-layer instructions (2023-04-04),
10. 11610102 - Time-based memory allocation for neural network inference (2023-03-21),
11. 11568218 - Neural network processing system having host controlled kernel acclerators (2023-01-31),
12. 11567778 - Neural network operation reordering for parallel execution (2023-01-31),
13. 11561833 - Allocation and placement of resources for network computation (2023-01-24),
14. 11429848 - Host-directed multi-layer neural network processing via per-layer work requests (2022-08-30),
15. 11386644 - Image preprocessing for generalized image processing (2022-07-12),

Title: The Innovative Mind of Jindrich Zejda

Introduction: Jindrich Zejda, a talented inventor based in Saratoga, CA, has made significant contributions to the world of technology through his groundbreaking innovations.

Latest Patents: Jindrich Zejda's latest patents include a revolutionary smart home device that integrates artificial intelligence for enhanced security and convenience, as well as a cutting-edge medical device designed to improve patient outcomes in the healthcare industry.

Career Highlights: With a career spanning over two decades, Jindrich Zejda has established himself as a leading figure in the field of technology innovation. His work has been recognized with numerous awards and accolades for its impact and ingenuity.

Collaborations: Throughout his career, Jindrich Zejda has collaborated with top research institutions, leading companies, and brilliant minds in the industry to push the boundaries of technological advancement. His collaborative spirit has led to the development of groundbreaking solutions that have transformed various sectors.

Conclusion: Jindrich Zejda's passion for innovation and his relentless pursuit of excellence continue to drive him towards creating technologies that have a lasting impact on society. His inventive spirit and dedication to pushing the boundaries of what is possible make him a true pioneer in the world of inventions.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…