Altera Introduces Agilex 9 Direct RF-Series SoC FPGA
ESD eBook June 2026: Building Blocks of Robotics Design
AI Agents at the Edge, Robotics, Motor Control ICs: Embedded Week Insights
QSPICE’s Native Python Socket API: Real-Time Co-Simulation for Power, Analog and Mixed-Signal Designers
For the first time, an Earth observation satellite has autonomously identified areas of interest using a vision-language model (VLM) without human analysts on the ground. The milestone occurred in April aboard Loft Orbital’s YAM-9 spacecraft, launched in fall 2025. A software package built by NASA’s Jet Propulsion Laboratory, called NAVI-Orbital, harnessed Google DeepMind’s Gemma 3 VLM—a model designed for edge applications on limited hardware. The VLM combines contextual understanding with image analysis, responding to natural language queries such as classifying intersections of natural and human environments or identifying infrastructure near railway hubs.
The demonstration marks the first reported use of a VLM in orbit. In the near term, it enables satellites to perform initial data triage, reducing the flood of raw data sent to Earth. Longer term, it proves the feasibility of running larger-scale AI in space, potentially enabling "always-on patrol layers" that monitor borders or detect suspicious activity. Loft Orbital’s head of AI, Paul Lasserre, emphasized the capability for interactive logic with satellites.
Loft’s spacecraft serve as platforms for third-party customers, akin to infrastructure-as-a-service. YAM-9 includes an Nvidia Jetson Orin AGX GPU, a leading chip for space compute. NAVI-Orbital’s development was led by NASA JPL’s Juan Delfa Victoria, who streamlined Gemma 3’s libraries and memory for orbital use.
Other companies are following suit: Planet Labs flies Jetson Orin processors for object detection and researches VLMs; Kepler Communications, with the largest space GPU fleet, notes undisclosed compute use cases since January. Loft aims to build a constellation of 50–100 such satellites for real-time global coverage (currently 12 in orbit). Lessons from small-model deployment will inform larger-scale space compute, particularly in power and memory management. The concept originated from JPL’s work on digital assistants for astronauts exploring the moon or Mars, providing an interactive AI akin to video game assistants.
The AMD Radeon RX 9070 GRE, originally launched as a China-only GPU last year, received a global release at Computex 2026. It targets the upper-mainstream 1440p gaming segment, positioned below the RX 9070. Built on RDNA 4 architecture, it features 48 compute units, 48 ray accelerators, 96 AI accelerators, and an enhanced media engine supporting AV1 encode/decode. It comes with 12GB GDDR6 on a 192-bit bus (432 GB/s bandwidth) and a 220W TBP, requiring a 650W PSU. The card uses two 8-pin PCIe connectors and includes DisplayPort 2.1a and HDMI 2.1b outputs.
AMD claims the RX 9070 GRE outperforms the GeForce RTX 5060 Ti 16GB by up to 22% across raster and ray-tracing games and offers 26% better performance per dollar. However, at $549 MSRP, it matches the original launch price of the RX 9070 (now $619), which has 16GB VRAM and a wider 256-bit bus, highlighting current memory costs rather than the GRE's standalone value. For gamers upgrading from older 1080p or early 1440p GPUs, it offers a full-platform upgrade with PCIe 5.0 and modern features.
The RX 9070 GRE is an AIB-only card; partners include Acer, ASUS, ASRock, Gigabyte, PowerColor, Sapphire, and XFX. The review sample (PowerColor) uses a triple-fan cooler with a metal backplate and ventilation. In UL Procyon AI text generation benchmarks, the GRE scored 1,579 (Phi), 1,699 (Mistral), and 1,526 (Llama3), ahead of the RX 9060 XT but behind the RX 9070 and RTX 5060 Ti. The 12GB memory may become a limitation in future heavy titles, but the card is primarily aimed at high-quality 1440p gaming with solid ray tracing and upscaling support.
Just when I think I’ve seen everything, I find myself being introduced to an AI-powered hardware-in-the-loop (HIL) testing framework that lets any embedded design team build a fully automated pipeline on real hardware in a matter of hours. I hail from the land before time and the days before embedded systems. This isn’t to say that embedded systems didn’t exist; it’s just that no one had coined the “embedded” moniker at that time, and the systems themselves bore little resemblance to what we think of as embedded systems today. Just to set the scene… how would you define an embedded system? Does it have to be electronic, or could it be clockwork? Must it be digital, or can it be analog? Does it have to involve a computer in the form of a microprocessor or a microcontroller? I’m afraid this is one of my hobby horses; that is, cogitating, ruminating, and waffling on about this with my embedded engineering friends is one of my favorite pastimes. For example, this is the point in the conversation where I would typically say something like, “But what about centrifugal governors, examples of which were invented by Christiaan Huygens in the seventeenth century?” I would then go on to note (irrespective of how hard they tried to stop me) that the same technology was used by James Watt to control steam engines. An example is shown below. As the speed of rotation increases, the two balls move outwards and upwards, thereby controlling a valve. By automatically adjusting the steam flow in response to engine speed, the governor formed a closed feedback loop that maintained a constant speed without human intervention. In 1868, James Clerk Maxwell analyzed such governors in his landmark paper On Governors, laying much of the mathematical foundation for modern control theory. Boulton & Watt steam engine circa 1788 (Source: Dr. Mirko Junge/Wikipedia) The point here is that this is purely analog, with no digital processing involved, yet I would argue that as a key element of a much larger system, it was the equivalent of an embedded controller—albeit one constructed entirely from brass, steel, and cleverness. Precursors to what we now consider to be embedded systems date back to the 1960s. The Apollo Guidance Computer (1965) is often cited as one of the first recognizably modern examples. However, engineers in the 1960s and much of the 1970s didn’t generally talk about “embedded systems.” They were more likely to refer to dedicated computers, process-control computers, microprocessor-based controllers, intelligent instruments, microcomputer-based products, and firmware-controlled devices. The phrase “embedded system“ didn’t start to become common until the late 1970s and early 1980s as microprocessors and microcontrollers began to find their way into industrial equipment, automobiles, appliances, instrumentation, and communications products. By the mid-to-late 1980s, this term was well established in the engineering press, textbooks, and conferences. For much of their history, embedded systems lived quiet, self-contained lives. They were tucked away inside washing machines, industrial controllers, automobiles, medical instruments, and countless other products. They monitored sensors, drove actuators, and faithfully performed their assigned tasks, but they rarely communicated beyond the boundaries of the system they inhabited. In fact, a definition of an embedded system that resonated with me at that time was: “A system you don’t even know is there… until it stops working.” Then came the Internet, followed by the Internet of Things (IoT). Suddenly, embedded systems were no longer isolated. They started talking to cloud services, smartphones, other devices, and each other. Data flowed in every direction. Updates could be delivered remotely. Intelligence could be distributed across entire fleets of devices. At the same time, many embedded systems emerged from their dark and dingy hiding places and began basking in the light of day. No longer buried deep inside industrial equipment, they appeared in smartwatches, fitness trackers, augmented-reality glasses, handheld medical devices, smart speakers, and countless other products that users interact with directly every day. In a sense, the term “embedded system” has become a victim of its own success. Many of today’s systems are no longer hidden away. They’re right there in front of us. Perhaps “in-your-face systems” would be a more appropriate description. Of course, if that term had caught on, generations of engineers would now be proudly describing themselves as “in-your-face system designers,” which somehow doesn’t have quite the same ring to it, but—as usual—we digress… The point is that embedded systems have reinvented themselves several times already. They evolved from mechanical analog controllers to digital microprocessor-based systems, from isolated devices to connected IoT nodes, and from hidden components to products in their own right. Now they appear to be embarking on yet another transformation—one in which artificial intelligence is becoming an active participant in the design, development, testing, and maintenance process itself. The reason for this historical ramble is that I was just chatting with Noah Pacik-Nelson, Co-Founder and CEO at BootLoop, along with Ollie Rubens, who is helping build out the company and bring its technology to larger enterprise customers. BootLoop may be a newcomer, having been incorporated only in the summer of 2025, but it has already attracted customers working on everything from aerospace systems and fusion-energy projects to medical devices, wearables, and datacenter infrastructure. Now, I don’t care what anyone says; that’s impressive! The founders bring a rather eclectic mix of expertise to the table. Noah’s background spans ultra-low-power AI research, embedded vision and speech systems, and large-language-model agent frameworks. The company’s other founder, Chris Markus, contributes the sort of firmware pedigree that tends to make embedded engineers sit up and take notice. Among other things, Chris served as the firmware lead for SpaceX’s Raptor engine and later led software efforts for Starship booster recovery. The two founders first worked together during the COVID-19 pandemic on a ventilator project and later collaborated at the MIT Media Lab before deciding to tackle what they saw as some of the biggest bottlenecks in embedded hardware development. The interesting thing is that BootLoop isn’t simply another AI coding assistant. The company views embedded development as a three-stage process: build it, test it, and fix it. To address these stages, the platform comprises three complementary offerings: Develop, Test, and Debug. The first of these, BootLoop Develop, focuses on firmware creation and hardware bring-up. Before generating a single line of code, the system ingests schematics, netlists, board design files, datasheets, reference manuals, register maps, API documentation, and other design collateral. In effect, it constructs a firmware-centric understanding of the hardware platform. Only then does it begin generating code. More importantly, it doesn’t simply generate code and wish the developer good luck. The generated firmware is continuously exercised on real hardware throughout the process. As the folks at BootLoop like to say, the goal is to create code that works, not merely code that compiles. BootLoop Develop can be used for new-board bring-up, driver development, MCU migration projects, feature additions, performance tuning, and security hardening. In each case, the system leverages its understanding of both the hardware and software environment to accelerate work that would traditionally consume days or weeks of engineering effort. At the opposite end of the lifecycle lies BootLoop Debug, internally known as Sentinel. This tool addresses the inevitable reality that no matter how carefully we design our systems, bugs will eventually escape into the wild. When a bug report arrives—perhaps as a Jira ticket accompanied by logs, crash dumps, or diagnostic data—Sentinel analyzes the available information alongside the codebase itself. It then generates a root cause analysis (RCA) report that highlights likely failure mechanisms, relevant files, execution paths, and potential fixes. Over time, the system develops an institutional memory of previous failures and their resolutions, allowing it to recognize recurring patterns and accelerate future investigations. Although both Develop and Debug are impressive in their own right, the real reason for our conversation was the launch of BootLoop Test, which occupies the critical middle ground between firmware creation and field maintenance. BootLoop Test: From zero to CI in under 30 minutes (Source: BootLoop) To understand why this matters, consider the current state of hardware-in-the-loop (HIL) testing. The most sophisticated hardware companies have long relied on HIL environments to validate their products. Unfortunately, creating and maintaining these environments often requires months of specialized work. As a result, many engineering teams continue to rely on manual bench testing, collections of one-off scripts, and procedures that exist only inside the heads of a handful of experienced engineers. BootLoop’s ambition is to make enterprise-grade HIL testing as easy to deploy and scale as modern continuous integration (CI) systems. At the heart of the approach is the same hardware understanding that powers the BootLoop Develop product. Because the platform already understands the schematic, signal paths, peripherals, register maps, and their relationships, it can reason about the hardware in a surprisingly sophisticated fashion. If a peripheral fails to initialize correctly, for example, the system can inspect debugger information, interact with GDB (GNU Debugger), examine register contents, monitor serial interfaces, and correlate those observations with its knowledge of
The STH Newsletter is a weekly Saturday digest curated by Patrick Kennedy, designed for readers who lack time to visit the site daily. It highlights key market coverage, reviews, and industry events from the past week. The newsletter is personally written by Kennedy each week, even while traveling, and is sent directly to subscribers’ inboxes.
In Q3 2024, STH launched a paid Substack called the Axautik Group (pronounced “exotic”), which serves as the analyst arm of STH. This second publication provides more detailed content not suitable for the main STH site, targeting financial communities and data center infrastructure providers. It also offers a way for readers to support STH’s work. A sample paid report on the Substack compares the Qualcomm Cloud AI 100 with LPDDR options from AMD, NVIDIA, and Apple.
The newsletter format has settled into three sections: a recap of the week’s biggest industry-impacting topics, the top five stories chosen by the STH team (focusing on reviews and personal favorites, free from advertising influence), and a preview of upcoming content and major industry events.
Subscribing is simple via a form on MailChimp, with no email selling and easy unsubscribe options. Kennedy emphasizes that the newsletter is designed to be a useful, burden-free tool for readers. To maintain a positive experience, STH avoids newsletter signup overlays and pop-ups, running the site as Kennedy himself would want to visit it. The goal is to help readers quickly catch up on STH content even if they miss daily visits.
ASRockRack NUC BOX 358H Front Angled 1 While the next unit of computing (NUC) form factor has lost some of its steam since its chief supporter, Intel, divested itself of its in-house NUC offerings a few years back, the ultra-compact form factor has held on thanks to its unique size. At roughly half the volume of even a 1-liter mini-PC, the 4-inch-by-4-inch boxes have proved incredibly handy for computer installations that require a truly miniature, out-of-the-way PC, leading companies such as ASRock Industrial to continue developing new systems even after Intel divested the business. If anything, these days ASRock Industrial is one of the most ardent and consistent supporters of the platform, having released NUCs based on Intel and AMD hardware over several generations. Today, we are looking at the latest-generation NUC, the BOX-358H, which is built around Intel’s Panther Lake Platform. ASRock Industrial NUC BOX-358H Key Specs Processors Intel Core Ultra X7 358H, 4P + 8E + 4 LPE (4.8GHz) Operating System N/A Memory As Shipped: Empty (2x SO-DIMM) As Tested: 96GB DDR5-5600 (2x48GB SO-DIMM) Storage As Shipped: Empty M.2 2280 (PCIe Gen5 x4) + M.2 2242 (PCie Gen4 x4) As Tested: 1TB Crucial P3 Plus (PCIe Gen4 x4) GPU Intel Arc B390 (Xe3, 12 Xe cores) Form Factor NUC Dimensions 117.5 x 110 x 49 mm (4.63 x 4.33 x 1.93 in) Weight 1 kg (2.2 lbs) Wireless Wi-Fi 7 + Bluetooth 6.0 (Intel BE211) Color Black Ports Front: 1x USB-C/Thunderbolt 4 40Gbps w/DP Alt Mode, 1x USB-C 20Gbps, 1x USB-A 10Gbps, 1x Combo Audio Jack Rear: 2x USB-A 10Gbps, 2x HDMI 2.1, 1x 2.5GbE (I226LM), 1x 2.5GbE (I226V) Under the hood of the little box that would make even most TinyMiniMicro PCs blush, ASRock Industrial has incorporated Intel’s latest generation laptop hardware, culminating in a barebones box that can rival a high-end ultraportable PC. With the Core Ultra X7 358H sporting a full-fat Panther Lake implementation, including 16 CPU cores and Intel’s best integrated GPU, the BOX-358H is deceptively powerful for its size. Especially with the Panther Lake platform bringing numerous upgrades to Intel’s laptop hardware, the BOX-358H represents a significant performance step up over ASRock Industrial’s earlier NUC boxes. VIDEO If you wanted to find the ASRock Industrial NUC BOX-358H online, here is a Newegg affiliate link. ASRock Industrial NUC BOX-358H External Hardware Overview If there’s one thing to be said about ASRock Industrial’s NUC offerings, it is that the company is remarkably consistent both in offering a new NUC with each generation of hardware from Intel and in the design of those NUCs. We have been reviewing NUC boxes from the company since early on in the decade, and other than replacing a glossy chassis with a matte one a couple of years back, the itty-bitty PCs are practically unchanged from the outside. This is ultimately a fancy way of saying that if you have seen one, you have seen them all. It also means that, for ASRock Industrial’s customers, their NUCs can be quickly swapped and upgraded, allowing newer boxes to replace older ones. Intended first and foremost as an industrial PC, ASRock Industrial is not trying to impress anyone with its looks here. It is all about the functionality. ASRock Industrial NUC BOX 358H Front 1 Starting our brief tour from the front of the system, we have ASRock Industrial’s standard NUC front port layout. This includes a 10Gbps USB-A port, a 20Gbps USB-C port, a 40Gbps USB-C/Thunderbolt 4 port, and finally, a combo audio jack. This is the bulk of the system’s USB I/O right here, with the TB4 port offering the greatest bandwidth and the greatest flexibility in what can be connected to it. Try to guess which is which based on the labeling. ASRock Industrial NUC BOX-358H USB-C Ports. Left: TB4/USB4, Right: USB3 This single TB4 port makes ASRock Industrial’s box unusually notable in that it remains one of the only systems we have ever come across with a single TB4 port. Intel’s SoCs natively support TB4 ports in pairs (or better). Virtually every other vendor installs multiple TB4 ports. As ASRock Industrial opted to integrate only a single TB4/USB4 retimer inside their PC, the system can only offer TB4 on a single USB-C port, the left one. Meanwhile, the right USB-C port is instead limited to 20Gbps USB3 (Gen 2×2) speeds. Both of these USB-C ports are DisplayPort Alt Mode capable as well, should the need arise. The left port can support DP 2.1 speeds, while the right port supports DP 1.4. ASRock Industrial NUC BOX-358H Side 2 A quick look at the left side of the NUC shows that ASRock Industrial uses both sides for ventilation. There is no front air intake of any kind, so these vents are to enable passive airflow for cooling less critical parts of the system. ASRock Industrial NUC BOX-358H Rear 1 Meanwhile, at the rear of the box, we find the rest of the system’s I/O ports. Here, ASRock Industrial has placed two more 10Gbps USB-A ports, two HDMI 2.1 ports, and two 2.5GbE ports. One is driven by Intel’s vPro-capable i226LM controller, and the other is driven by Intel’s i226V controller. Combined with the DP Alt Mode capabilities of the front USB-C ports, the NUC BOX-358H can drive up to 4 displays in general. While the system’s power consumption is low enough to be powered by a USB-C power adapter, in keeping with both the design consistency of the NUC BOX family and industrial customers’ needs, ASRock Industrial has stuck with a DC barrel connector. A rather flexible one at that, with the system capable of handling input voltages ranging from 12V to 24V, allowing it to be used with a wide variety of power supplies. Unfortunately, none of this is very well labeled. Presumably, in keeping with the generic, reusable design of ASRock Industrial’s NUC chassis, none of these USB, HDMI, or Ethernet ports are labeled with their speeds. Instead, everything is simply labeled with the type of port it is, information that anyone familiar with a PC would easily recognize. So there is a missed opportunity here from ASRock Industrial to better label their ports with their technical capabilities, which, unlike the port type, is non-obvious information. ASRock Industrial NUC BOX-358H USB 3.2 Gen2 Type A Ports 2 Ports aside, the back side of the system is also where the NUC BOX-358H’s sole exhaust vent is. As we will see a bit later, this is driven by a single blower fan that cools the Intel SoC. Moving on, a quick look at the top of the system reveals a border with ventilation holes, from which the system’s fan draws in fresh air. ASRock Industrial NUC BOX-358H Top 1 Finally, at the bottom of the system, we see the rubber feet that the system stands on, as well as the four screws underneath that keep the bottom plate firmly attached. ASRock Industrial NUC BOX-358H Bottom 1 Because this is a barebones system that does not ship with any configuration options or storage (let alone an operating system), ASRock Industrial has kept the system design very clean. Even on the bottom of the system, you will not find any stickers, serial numbers, or COAs. In fact, you will not find anything at all. Nowhere on the system has the company stenciled or printed the system’s model number or even the manufacturer’s brand. It is truly a nondescript black box. Out of the box, the system ships configured for desktop use. However, ASRock Industrial also offers a VESA mounting bracket for more secure (and vertical) installations. ASRock Industrial NUC BOX-358H Bracket 1 Before diving into the guts of the system, here is a quick look at the included power supply as well. ASRock Industrial NUC BOX-358H External Power Supply 1 ASRock Industrial ships the system with a rather large 120W (19V@6.32A) power supply, which is wider than the NUC itself. ASRock Industrial could presumably reduce the size of the included power adapter by using a GaN-based adapter, though for their target market, the size savings are likely not worth the cost. With the external tour complete, let us crack open the BOX-358H and take a look inside.
Aewin Technologies, a member of the BenQ Qisda Group, showcased its next-generation network appliances and servers at Computex 2026, held from June 2 to June 5 at the Taipei Nangang Exhibition Center (booth M0104). The exhibition highlighted the convergence of high-performance computing and advanced networking architectures for AI-driven applications, with platforms including Taiwan Excellence Award-winning HA storage servers, edge computing platforms, general-purpose servers, and high-performance network appliances designed to accelerate AI-powered cybersecurity, Edge AI, and smart storage deployments.
“As Agentic AI is rapidly reshaping cybersecurity and networking environments, the demand for infrastructure capable of real-time analytics, large-scale data processing, high-throughput networking, and accelerated storage performance is growing exponentially,” said Charles Lin, GM of Aewin Technologies. He emphasized deep collaboration with Intel and AMD to deliver reliable, performance-optimized platforms for modernizing data center architectures.
Key products introduced include the MIS-5131 2U2N HA storage server powered by Intel Xeon 6 processors, supporting up to 24 dual-port NVMe SSDs with BMC heartbeat monitoring and NTB inter-node communication for real-time failover. The SCB-1953 combines Intel Xeon 6 processors, QAT, and PCIe Gen5 with Intel E830 NICs to enhance packet processing and threat detection. The dual-socket BIS-5231 general-purpose server features two Intel Xeon 6 CPUs, 12 3.5″ drive bays, and eight PCIe Gen5 slots for flexible deployment of NICs, GPUs, FPGAs, and encryption accelerators. The BAS-6101, powered by AMD EPYC 9005 processors, offers eight PCIe Gen5 slots and one OCP 3.0 slot in a short-depth chassis for space-constrained edge computing.
Aewin’s comprehensive portfolio integrates high-performance computing, networking, and cybersecurity technologies to deliver secure, scalable, and low-latency platforms for diverse AI applications across edge, enterprise, and data center environments. The company will continue strengthening ecosystem partnerships to accelerate AI innovation and adoption worldwide.
AI Agents at the Edge, Robotics, Motor Control ICs: Embedded Week Insights Here’s a roundup of this week’s must-read articles. We’ll look into the latest developments on agentic AI at the edge, robotics, and motor control ICs. Also, check News Archives – Embedded and Technical Articles Archives – Embedded for the complete list of news and articles from our website. NEWS The industry-first validation framework helps ensure interoperability, functional safety, and faster deployment of A-PHY systems. TÜV NORD-certified development processes cover coding guidelines, vulnerability management, and patch handling for embedded building blocks and tech stacks. The interactive simulation platform helps engineers evaluate switching behavior and reduce design iterations early. Avocado OS, an integrated hardware and software platform, helps developers scale vision AI systems from prototype to deployment. You can find more news here: News Archives, Products Archives ARTICLES Nvidia unveils a new Arm-based superchip for agentic AI on PCs that could be the basis for future embedded processors. In robotic systems, a real-time operating system (RTOS) offers high predictability, higher security, and more reliable safety systems. Motor control ICs designed for high performance power the next-generation of robotic applications. Chiplets and heterogeneous computing meet the complex design challenges of robotics development. Embedded autonomy stacks combine data from multiple sensors to improve perception, localization, and navigation reliability. You can find more articles here: Technical Articles Archives – Embedded, Analysis Archives – Embedded
Kontron has introduced the VX33211, a high-performance 3U VPX graphics processing board designed to bring advanced AI acceleration, graphics processing, and parallel computing to rugged embedded systems. Targeting defense, aerospace, and other mission-critical applications, the board combines Nvidia’s latest Blackwell GPU technology with an open-architecture approach for demanding edge-computing workloads.
At its core is the Nvidia RTX PRO 2000 Blackwell Embedded GPU, delivering up to 13.78 TFLOPS of FP32 performance. It integrates 3,328 CUDA cores, 104 Tensor Cores for AI inference and machine learning, and 26 RT Cores for real-time ray tracing and advanced visualization. The board includes 8 GB of GDDR7 memory with bandwidth up to 384 GB/s, enabling real-time execution of complex algorithms at the edge while reducing latency and reliance on centralized resources.
The VX33211 is built for harsh environments. The conduction-cooled version complies with VITA 48 and meets VITA 47 environmental standards, operating from –40°C to 85°C at the card edge. An air-cooled version is available for development and lab use. Designed around SOSA-aligned OpenVPX profiles, it promotes interoperability and simplifies integration into existing defense and aerospace architectures. PCIe Gen4 connectivity runs through the VPX backplane, and an integrated IPMI controller supports VITA 46.11 Tier 3 functionality for remote monitoring and out-of-band management.
The board complements Kontron’s VPX ecosystem, including CPU platforms like the VX307C, enabling balanced heterogeneous computing architectures. It is also supported in the HARAKAN-F rugged computing platform. Typical applications include intelligence, surveillance and reconnaissance (ISR), electro-optical and infrared processing, radar and sonar analysis, electronic warfare, mission computing, visualization systems, and embedded simulation and training. By integrating Nvidia’s Blackwell architecture into a rugged VPX form factor, the VX33211 offers a scalable platform for next-generation edge AI and high-performance embedded computing.
Congatec Earns IEC 62443-4-1 for Embedded Development Processes Congatec has received IEC 62443-4-1:2018 certification, issued by TÜV NORD, covering the development and support of its embedded building blocks and technology stacks. The certification validates that cybersecurity requirements are systematically integrated across all phases of the development lifecycle, from initial design through ongoing maintenance and obsolescence management. The certified scope includes congatec’s application-ready building blocks and its aReady.COM technology stacks, which combine computer-on-modules with licensed operating systems, including Ubuntu Pro and ctrlX OS, as well as software components such as conga-connect (aReady.IOT) for IoT connectivity and conga-zones (aReady.VT) for virtualization. Certified processes cover secure coding guidelines, verification and validation procedures, and structured management of vulnerabilities, patches, and obsolescence. For OEMs and system integrators, the certification provides a documented, auditable foundation for building compliant embedded products. It simplifies security verification for both customers and regulatory authorities, reduces internal certification overhead, and strengthens supply chain transparency. These advantages are particularly relevant in regulated sectors such as industrial automation and robotics, medical technology, energy systems, and transportation. The certification also carries direct regulatory significance. The EU Cyber Resilience Act (CRA), which becomes mandatory on December 11, 2027, imposes cybersecurity obligations on manufacturers placing electronic products on the EU market. Customers building on congatec’s IEC 62443-4-1-certified tech stacks gain a structured head start in assembling the compliance and security evidence required under the CRA and comparable regulatory frameworks. By grounding its embedded portfolio in internationally recognized, audited development processes, congatec enables customers to accelerate product development without having to rebuild security foundations from scratch. Engineering teams can leverage pre-validated building blocks to shorten time to market, particularly in verticals where security certification accounts for a significant portion of the overall development effort. For further technical details, visit congatec’s webpages about their security-by-design approach and certified technology portfolio.