Klepsydra Technologies AG provides advanced, patented edge AI software solutions for industries such as aerospace, robotics, and automotive. We focus on developing software that maximizes the potential of AI and edge data processing to deliver high performance and efficiency.
Our mission is to offer scalable and efficient software that integrates seamlessly across a wide range of hardware, from cost-effective processors to high-performance accelerators. By addressing the balance between computational power and efficiency, we help our clients enhance their operations and achieve their objectives.
Klepsydra Technologies aims to be a global leader in AI and edge data processing solutions. Our partnerships with edge processor manufacturers and operating system providers strengthen this vision. Our software has also been successfully tested in space through multiple projects, demonstrating its reliability and innovation.
Our Products
Efficient, lightweight, modular, and compatible with most used operating systems.
Klepsydra’s acceleration technology combines two-dimensional parallelization with lock-free programming to deliver a flexible and efficient inference engine, achieving up to 4x lower latency, 10x higher throughput, and 50% reduced power consumption. Its autotuning tool optimizes AI inference for various constraints, while a unified API eliminates processor lock-in, enabling greater flexibility and improved performance across hardware platforms.
Klepsydra AI: High-Performance Edge AI Framework
Klepsydra AI is a cutting-edge software framework for Edge AI applications, delivering high-efficiency AI inference. It offers high performance, broad compatibility across edge computers and operating systems, and a lightweight, independent design with no third-party dependencies, making it a powerful and adaptable solution for next-generation edge computing.
Klepsydra AI – Space Edition
The Space Edition of Klepsydra AI is built for space applications, ensuring high reliability and performance in extreme environments. ECSS-qualified by ESA, it supports radiation-hardened processors used in satellites and spacecraft. Additionally, it is compatible with the RTEMS space-qualified operating system, ensuring seamless integration for critical missions.
Klepsydra AI – Robotics Edition
The Robotics Edition of the Klepsydra Software Framework is specifically designed for robotics applications, ensuring high efficiency and real-time AI inference. It features an advanced ROS2 Executor, optimizing performance for robotic systems by reducing latency and improving processing speed, making it ideal for autonomous robots, industrial automation, and AI-driven robotics.
AI Everywhere
Our unified API for AI acceleration provides a real vendor-independent solution, delivering enhanced performance and flexibility compared to open-source alternatives.
Easy-to-Use
Reduce costs and risks with a unified, hardware-independent API that avoids re-development when changing processors.
Performance
A flexible and efficient inference engine optimized for three key performance metrics: a 4x reduction in latency, a 10x increase in throughput, and a 50% reduction in power consumption.
Use Cases
Klepsydra excels in applications that demand the processing of large data volumes or the execution of processor-intensive algorithms. Key use cases include:
- Edge AI Applications: Ideal for automotive applications where numerous sensors generate massive data that must be processed on-board. It also supports IoT real-time applications, including cybersecurity, surveillance, and GPS-denied environments.
- Space Applications: Essential for on-board processing of Earth Observation data (e.g., early warning systems), Space Situational Awareness, space debris removal, and lunar landing missions.
- Robotics Applications: Perfect for autonomous robotic systems that rely on sensors for navigation, as well as real-time robotic applications that require deterministic, reliable software performance.
Klepsydra Explained
Struggling with edge AI implementation, limited resources, or performance bottlenecks? Klepsydra Technologies provides solutions to optimize your intelligent systems and streamline AI deployment at the edge.
Klepsydra’s acceleration technology combines two-dimensional parallelization with lock-free programming to deliver a flexible and efficient inference engine, achieving up to 4x lower latency, 10x higher throughput, and 50% reduced power consumption. Its autotuning tool optimizes AI inference for various constraints, while a unified API eliminates processor lock-in, enabling greater flexibility and improved performance across hardware platforms.
OUR DEMOS
AI-as-a-Service
Three AI algorithms running in parallel in the LX2160.
Lunar Landing on multiple space processors
Demonstrating Klepsydra's Versatility Across Multiple Space Processors
Data aggregation demo
This demo showcases data aggregation in resource-limited environments with multiple data sources.
MEET OUR TEAM
Dr Pablo Ghiglino
Founder & CEO
Pablo, the CEO and inventor of Klepsydra's technology, is a word-class software...
Isabel del Castillo
Co-Founder & COO
Isabel holds a Master's degree in Philosophy and brings over 20 years...
Klaus Buchheim
Head of Business Development
Klaus, with a background in Electrical Engineering and over 15 years of...
Mandar Harsche
Senior C++ developer
"Mandar Harshe, a Senior C++ Developer at Klepsydra Technologies holds a PhD...
Guillermo Sarabia
AI Software Engineer
Guillermo is a Senior AI Software Engineer at Klepsydra Technologies. He specializes...
Tomasz Drążek
DevOps Engineer
With an M.Sc. in Computer Science, Tomasz brings over 9 years of...
Dr Manuel López Palma
FPGA Engineer
With an MSc in Physics and a PhD in engineering, Manuel has...
Mario A. Rueda
Junior C++ Developer
With a BSC in Computer Science, Mario has 2 years of experience...
Jose Luis Barros Fernandez
C++ Developer
Jose Luis is a Software Engineer at Klepsydra Technologies, where he has...
Marc Mateu
C++ Developer
Marc is a C++ Developer at Klepsydra Technologies, where he has been...
Caio Toledo
C++ Software Engineer
With a Bachelors Degree in Electronics Eng. Caio has more than 10...
Thanos Andriopoulos
RTOS and C++ developer
Thanos has an MEng in Electrical & Computer Engineering, 4 years of...
Diego Sillero
Software Q&A
A versatile Software Quality Engineer with eight years of experience in critical...
Mario Hochreiter
Technical Advisor
Mario, an MSc in Computer Science, brings 20 years of software engineering...
Miguel Rodriguez
Project Leader
Miguel holds an MSc in Mechanical Engineering and has over 30 years...
Marcel Haltiner
CFO & Financial Advisor
With a Master's in Business Administration and over 15 years of finance...
Silvan Guler
Strategy & Investor Relations
Silvan joined Klepsydra Technologies in 2021 as an advisor and investor, focusing...
Klaus Decking
Board Member
With a German Mechanical Engineering degree and a US MBA, Klaus brings...
LATEST NEWS
Klepsydra AI Successfully Demonstrates In-Orbit Performance on Spire Global Satellite
(Image courtesy of Spire Global) Zug, November 2024 – Klepsydra Technologies is proud to announce the successful in-orbit demonstration of its cutting-edge AI inference engine on a Spire Global satellite. Conducted on the night of October 31, 2024, in collaboration with the European Space Agency (ESA) and Spire Global, this milestone showcases the exceptional performance… Continue reading Klepsydra AI Successfully Demonstrates In-Orbit Performance on Spire Global Satellite
SYSGO and Klepsydra Announce Strategic Partnership to Drive AI Technology in Avionics and Space Embedded Systems
SYSGO and Klepsydra Announce Strategic Partnership to Drive AI Technology in Avionics and Space Embedded Systems Klein-Winternheim, Germany, and Zurich, Switzerland, Nov, 2024 - SYSGO, a leading provider of real-time operating systems (RTOS) and hypervisors, and Klepsydra, a specialist in Artificial Intelligence (AI) software for Edge Computing, are excited to announce a technology partnership aimed… Continue reading SYSGO and Klepsydra Announce Strategic Partnership to Drive AI Technology in Avionics and Space Embedded Systems