Klepsydra™ Returns to Embedded World with Live Demos and Presentations!
We’re excited to be back at Embedded World this year!
Klepsydra is a scale-up company that helps businesses bring Artificial Intelligence and advanced data processing to real-world edge systems. Our solutions enable faster, safe, and secure AI applications to run directly on devices such as robots, autonomous systems, and aerospace platforms, in highly constrained environments.
At Embedded World, we’ll be showcasing live demos and giving presentations both at our own booth and at the booths of our partners, sharing the latest advances in embedded systems and edge AI.
We’re looking forward to great conversations, new collaborations, and another inspiring edition of Embedded World. See you there!
Our partners at embedded world
Klepsydra / Embedded Brains Booth
Klepsydra and Embedded Brains are collaborating to deliver a full safety-critical AI software stack for space applications. Together, they have enabled Klepsydra AI on RTEMS 6 SMP across leading space-grade platforms, including AMD UltraScale+, AMD ZedBoard, and Microchip PolarFire Icicle, supporting advanced use cases such as Earth observation and vision-based navigation.
WindRiver Demo
Klepsydra AI is officially supported on Wind River VxWorks, making it the first commercial multi-core AI inference engine for this OS.
Together, Klepsydra AI and VxWorks deliver high-performance AI deployment with industry-leading safety, security, and determinism.
Visit the Wind River and Klepsydra booths to see a live GPS-denied navigation demo on a Toradex platform.
Toradex Demo
Klepsydra AI is fully supported on Toradex Apalis modules and available through the Torizon Demo Gallery.
The trial version delivers state-of-the-art performance and can be freely downloaded and evaluated by requesting a trial license.
Klepsydra also supports the Toradex Aquila TI AM69 platform, enabling efficient CPU-based AI deployment. A future release will add native NPU acceleration, allowing seamless scaling from CPU-only to CPU–NPU execution.
Emproof Demo
Klepsydra and Emproof have partnered to provide the highest level of cybersecurity for Edge AI. By merging Klepsydra’s high-speed AI acceleration with Emproof’s hardened binary protection, they offer a uniquely secure and efficient software stack. This partnership will be showcased in a demo featuring autonomous vehicle mono-camera navigation, proving that real-time AI can be both lightning-fast and tamper-proof.
DDC-i Integration
Klepsydra AI and DEOS have developed a unified software architecture that provides a secure, reliable, and high-performance framework for deploying AI applications. It delivers a standalone AI engine with a secure SDK for integration into safety-critical systems, enables framework-agnostic execution for consistent Edge AI performance, and supports high-throughput, low-latency computation for real-time responsiveness. The architecture also exposes a hardware-independent layer for seamless deployment across diverse platforms and simplifies AI model updates and lifecycle management with minimal overhead..
Red-Pitaya Demo
Klepsydra AI supports Red Pitaya platforms, enabling high-performance execution of AI workloads on these devices.
In addition, Klepsydra AI is officially qualified by the European Space Agency (ESA) for safety-critical missions on ARMv7 architectures, meeting stringent reliability and robustness requirements.A health-monitoring demonstration, jointly developed with Fraunhofer, is available at both the Red Pitaya and Klepsydra booths, together with detailed technical information on this integrated solution..
Also Present
wolfSSL Booth
Klepsydra software is now powered by wolfSSL, enabling Klepsydra AI to run commercially on RTOS-based platforms such as RTEMS and VxWorks. The integration delivers lightweight, high-performance TLS 1.3 security for embedded and real-time systems. For more information, visit the wolfSSL booth.
AEI and Klepsydra Showcase Advanced Self-Driving Navigation Demo
AEI and Klepsydra present their most impressive demonstration yet at the Klepsydra booth. The demo features a UAV simulator powered by an NVIDIA board running four parallel AI algorithms. By accelerating in-vehicle sensor data processing, Klepsydra AI enables AEI’s vision-based navigation to operate reliably while providing a scalable interface for simultaneous execution of multiple AI algorithms on NVIDIA GPU edge devices.