’Tis the Season to Scale Your Jetson Fleet

Why Jetson projects stall at scale—and how Avocado OS helps teams deploy, provision, and upgrade fleets with confidence.

Published by Peridio

Peridio is an enterprise-grade platform that offers a unified solution for building, deploying, and managing embedded Linux devices and connected products.

41 Members0 Projects
Author: Peridio

The dust has officially settled from Embedded World North America and what a month it’s been.

Justin packed the room with his session on Physical AI, and our booth demo ran nonstop: real robotics workloads, real edge AI constraints, and Avocado OS running it all on production-grade NVIDIA Jetson. The reaction confirmed what we’ve been saying for months: Physical AI is here, but the underlying software stack hasn’t caught up.

That’s especially true in the Jetson ecosystem.

Across robotics, industrial systems, and computer vision deployments, we’re seeing the same story repeat:

  • Teams prototype fast on Ubuntu, then hit a wall in manufacturing.
  • Yocto gives you control, but maintaining it becomes a full-time job.
  • Scaling Jetson devices beyond the lab becomes an exercise in version roulette, manual provisioning, and crossed fingers.

It shouldn’t be that way.

With Avocado OS, we’re making Jetson deployable, not just usable. Deterministic images across Orin Nano → NX → AGX. JetPack version consistency you can trust across hundreds of units. Provisioning and configuration in under 60 seconds. And an upgrade story that doesn’t require rewinding your entire stack.

Physical AI deserves better infrastructure. And we’re building it.. starting first with NVIDIA Jetson (big announcement coming soon!).

Next stop for the team is CES. If you’re going to be there, I’d love to connect.

— Bill

Celebrating the hardware innovators turning prototype dreams into production realities

The top 5 SOMs making Physical AI merry and bright ✨

This holiday season, we’re highlighting the SoM providers and hardware platforms setting the pace for what’s possible at the edge.

Physical AI is exploding — and none of it happens without the right hardware. In our latest collaboration with Hackster, we rounded up the top system-on-modules powering real robotics, vision, and industrial AI workloads heading into 2026.

  • From Orin-powered acceleration to compute-ready AMR platforms
  • What’s driving developer adoption
  • Why SOMs (not SBCs) are becoming the foundation for scalable AI devices

Read the full article on Hackster.

Robotics, real workloads and a "Best in Show" win, all in one week!

ICYMI: Embedded World North America highlights

Embedded World NA was a milestone for the Avocado OS community — and the momentum from the show hasn’t slowed.

Here’s what you may have missed:

  • Avocado OS wins “Best in Show” (Operating System): Awarded by Embedded Computing Design — a huge validation of the work the team is doing to modernize embedded Linux for Physical AI.
  • Justin Schneck’s packed session on Physical AI: Justin breaks down why modern devices demand new OS architecture — and how Avocado OS gives teams the foundation they need for robotics, computer vision, and industrial AI at scale.
  • Our live robotics demo: Dual-arm pick-and-place, visual inspection, real-time computer vision — powered by Avocado OS, NVIDIA Jetson, Roboflow, Solo Tech, and Advantech.

Watch the session,check out the demo, or catch Justin's Avocado OS interview.

Are you pushing the boundaries in physical AI, robotics or intelligent systems? Be a guest on the show!

🎙️ Reinventing embedded hardware with Grinn

We sat down with Robert Otręba, Founder & CEO of Grinn, a Poland-based embedded engineering company operating for nearly 18 years. Robert shares how Grinn grew from a two-person startup into a 70-engineer team specializing in system-on-modules (SOMs), embedded Linux, and high-speed hardware design.

🎙️ How Vision-Language-Action models are redefining Robotics

Bill sits down with Dhruv Diddi, founder and CEO of Solo Tech, to explore how Vision-Language-Action (VLA) models are transforming robotics. Dhruv breaks down how these models enable robots to see, understand, and act, bridging AI and the physical world.

🎙️ The evolution of AI in microcontrollers and embedded systems

In this episode, we chat with Henrik Flodell, Senior Marketing Director at Alif Semiconductor, a company redefining embedded AI. They discuss how AI is evolving in the semiconductor industry and how Alif’s architecture - combining neural processing units (NPUs) with microcontroller cores, enables low-power, hardware-accelerated intelligence at the edge.

🎙️ RealSense’s spinout from Intel, $50M Series A, and NVIDIA Robotics Alliance

We sit down with Chris Matthieu, VP of Developer Ecosystem and Chief Hacker at RealSense, to discuss the company’s spinout from Intel and its recent $50 million Series A raise. They explore how RealSense balances developer creativity with production readiness, advances in robotics and AI, the power of open source, and the growing importance of domestic tech manufacturing.

Catch up on all the episodes here.

FirstBoot is your front-row seat to the future of embedded systems. If you like what you see, hit Subscribe and share this newsletter with a teammate who’s tired of old-school embedded pain points.

No lock-in. No fluff. Just real-world embedded innovation.

See you next month!

Published by Peridio

Peridio is an enterprise-grade platform that offers a unified solution for building, deploying, and managing embedded Linux devices and connected products.

41 Members0 Projects

Authored by Peridio

Latest articles
Sponsored articles
Related articles
Latest articles
Read more
Related articles