Cytrence Kiwi Powers Seamless Developer Experience in the DGX Spark Workshop at NVIDIA GTC 2026
Cytrence Kiwi enables seamless, laptop-based console access to DGX Spark, simplifying scalable, hands-on AI training at NVIDIA GTC 2026
Release Date: March 30, 2026
Location: Burlington, MA
Burlington, MA – March 18, 2026 – Cytrence Technologies, maker of Kiwi KVM, collaborated with the NVIDIA team to deliver a successful “Hands-on DGX Spark Training Lab” during the NVIDIA GTC 2026 event at the San Jose Convention Center in California, USA.
See Kiwi introduction at 10m22s in the session recording on Nvidia website:
https://www.nvidia.com/en-us/on-demand/session/gtc26-dlit82308/
Bringing Desktop AI to Life
Since its introduction in October 2025, DGX Spark has redefined what is possible for desktop AI computing. By enabling developers to run advanced AI models locally, directly on their desktops, it offers significant advantages in data security, privacy control, and protection of domain-specific knowledge. As highlighted in NVIDIA’s official communications, “DGX Spark Ignites a New Era of Desktop AI.”
However, enabling that vision in a hands-on training sessions introduces practical challenges.
The Challenge of Training at Scale
While DGX Spark supports headless operation and remote access through a 10GbE Ethernet connection, initial setup and troubleshooting still require direct console access. The workshop was designed for 75 participants per session. Providing each attendee with a dedicated monitor, keyboard, and mouse is impractical due to space and budget constraints. At the same time, relying solely on network-based access in a hotel conference room where the session is hosted introduces friction and uncertainty. Also, given that participants have diverse technical backgrounds, reliable and instant access to DGX Spark is essential.
A New Approach to Accessing DGX Spark
To address these gaps, the workshop integrated Cytrence Kiwi, a compact USB 3.0-based KVM solution that transforms a standard laptop into a full console interface.
With Kiwi:
• HDMI output from DGX Spark is captured and displayed in a window on the laptop
• Keyboard and mouse input are passed directly from the laptop to DGX Spark
• No additional peripherals or network configuration are required
This approach enables each participant to interact with DGX Spark as if it were directly connected, using only a laptop supplied by the organizer. The result is a streamlined, distraction-free workflow:
• No external monitors
• No separate keyboard or mouse
• No dependency on network setup
Participants were able to focus entirely on the workshop content while accessing DGX Spark with native performance. As one attendee noted: “It’s really agnostic—almost magical—to access DGX Spark this way.”
Enabling the Next Generation of AI Workflows
Behind the scenes, the DGX Spark workshop team evaluated multiple approaches before selecting Kiwi for its simplicity and reliability. The solution integrates cleanly into diverse environments, making it well-suited for both structured training sessions and flexible development workflows.
This deployment highlights a broader shift in how developers interact with powerful computing systems. As AI moves closer to the desktop, tools that simplify access without adding complexity become increasingly important. By removing the need for traditional peripherals and reducing setup overhead, Kiwi helps bridge the gap between high-performance hardware and everyday usability.
Looking Ahead
The collaboration between Cytrence (a member of the NVIDIA Inception Program) and NVIDIA’s DGX Spark team demonstrates how thoughtful integration can elevate the developer experience at scale. By enabling seamless, direct interaction with DGX Spark systems in a classroom environment, Cytrence Kiwi helped make advanced AI workflows more accessible, efficient, and intuitive.
Cytrence looks forward to continuing to support the growing AI developer community and to unlocking new possibilities in desktop AI computing.
About Cytrence
Cytrence develops secure and reliable solutions for managing diverse computing devices for the future of enterprise IT, edge AI, and robotic engineering. Leveraging full-stack ownership and a security-first approach, Cytrence enables organizations to simplify workflows, improve efficiency, and strengthen operational excellence. The company is committed to advancing the future of device management and automation.
References
[1] https://blogs.nvidia.com/blog/live-dgx-spark-delivery/
[2] https://developer.nvidia.com/blog/new-software-and-model-optimizations-supercharge-nvidia-dgx-spark/
[3] https://blogs.nvidia.com/blog/dgx-spark-and-station-open-source-frontier-models/
Media Contact: media @ cytrence.com