Retronix Technologies Inc. announced the launch of two cutting-edge AI edge computing platforms, developed in collaboration with Renesas Electronics Corporation.The newly unveiled Sparrow Hawk Single Board Computer (SBC) and Raptor System on Module (SoM) are both powered by the latest Renesas R-Car V4H System-on-Chip (SoC), delivering up to 30 TOPS (Dense) of AI inference performance. These open platforms are designed to support a wide range of embedded edge AI applications and smart automotive solutions.Sparrow Hawk focuses on robotics, industrial automation, and rapid prototyping, offering a highly flexible and cost-effective development platform. Raptor, with its modular design and multi-camera processing capabilities, is engineered for commercial vehicles, advanced driver-assistance systems (ADAS), and autonomous guided vehicles (AGVs), meeting demanding requirements for reliability and AI edge computing.Product Highlight 1: Sparrow Hawk — A Versatile Platform for Edge AI ApplicationsSparrow Hawk is a compact and highly expandable edge AI development board featuring the Renesas R-Car V4H SoC. It offers up to 30 TOPS of dense AI inference performance and supports a fully open-source Linux environment, accelerating the development of industrial and embedded AI solutions.Key Features:*Optimized for Edge Intelligence: Designed for industrial robots, smart manufacturing, and autonomous control systems. *Raspberry Pi HAT Compatible: Easily integrates with popular modules and sensors to streamline development.*High AI Performance: Handles real-time image processing and AI workloads with ease thanks to 30 TOPS deep learning capabilities.*Open Development Environment: Built on an open-source Linux architecture with extensive community support.*Developer-Friendly Pricing: Campaign program at only USD 300, no paper contract required to get started.*Compact Design: Measures just 146mm x 90mm, ideal for embedded and terminal devices.Rich I/O and Expansion Interfaces:*8GB / 16GB LPDDR5 memory*Dual-camera interface and 40-pin GPIO header*1x DisplayPort, PCIe (4x USB3.0, 1x M.2 Key-M), 2x CAN-FD, Audio (2x In. 1x Out) and AVB Ethernet*Supports USB PD 20V power input and MicroSD removable storageRetronix Sparrow HawkPhoto: CompanyProduct Highlight 2: Raptor — Automotive-Grade AI SoM for Smart Vehicle Vision ProcessingRaptor is a high-performance SoM designed for automotive vision processing and edge AI computation. Powered by the Renesas R-Car V4H SoC, it supports multiple camera inputs, pre-processing, and AI inference. Raptor is ideal for applications including ADAS, smart cockpits, surround-view systems, and AGVs.Key Features: *Automotive-Grade Architecture: Built with safety-oriented design principles, long-term supply, and compliance with automotive standards.*Multi-Camera Support: Integrated ISP supports up to 8 video channels with synchronized vision processing.*Powerful AI and Specialized Automotive IP: Delivers 30 TOPS AI inference performance with integrated Image Rendering Unit, Dense Optical Flow, Structure from Motion, and CV/Deep Learning Engines.*Reference Carrier Design & Custom Development: Retronix offers reference designs and engineering services to accelerate product development. *Comprehensive Software Resources: Compatible with Yocto Linux and includes the Renesas AI Hybrid Compiler toolkit.*High Reliability: Designed for high-temperature environments with optimized power efficiency for automotive use.Retronix RaptorPhoto: CompanyAvailability and Computex ShowcaseSparrow Hawk and Raptor are scheduled to sample in late Q2 2025, alongside the launch of a developer program and open-source community support platform to help users rapidly prototype and deploy AI applications.We warmly invite industry professionals to visit Retronix at Computex 2025 (Booth N0814 / B-4) to experience the capabilities of Sparrow Hawk and Raptor firsthand. Explore their architectures, image processing performance, and AI inference efficiency across smart manufacturing, robotics, unmanned vehicles, and intelligent automotive applications.
The global wave of Artificial Intelligence (AI) is reshaping industrial paradigms across sectors. The convergence of wireless communication and AI is propelling AIoT into a new era, where diverse intelligent applications demand optimal performance, power efficiency, and cost-effectiveness to achieve scalable deployment. As a global leading provider of wireless communication modules and AI solutions, Fibocom is redefining its core competencies through an AI-centric platform approach, driving the deep integration of AI and connectivity to unlock new opportunities for intelligent transformation across industries."Traditionally, the core value of communication modules was 'connectivity,' but today, we're seeing AI capabilities rapidly extending to the edge," said Tiger Ying, CEO of Fibocom. "Modules are evolving into intelligent hubs capable of computation, sensing, and decision-making. The edge-cloud collaboration exemplifies the value of AI-communication convergence, and the role of the module is being redefined."Seamless Integration of Hardware and Software: Fibocom's Full-Stack Solutions Empowering the AIoT EraWith deep expertise in wireless communication, AI technologies, and integrated software hardware abilities, Fibocom is actively deploying AI at edge computing nodes to enable intelligent terminals. Through its "AI for X" initiative, the company delivers all-around AI capabilities, products, industry solutions, and ecosystem collaborations to accelerate the industry's transformation from "connect everything" to "intelligently connectivity."Fibocom's AI offerings are tailored for a wide range of vertical scenarios, including smart robotics, autonomous driving, industrial control, and smart retail. For lightweight AI applications, the company has introduced large-model solutions supporting voice and vision interaction, ideal for upgrading devices like AI toys and smart speakers. On the edge AI front, solutions such as the Nebula Series and 5G FWA SkyEngine Solution enable the deployment of large models directly on devices, significantly reducing latency and power consumption while enhancing real-time responsiveness and user experience.Fibocom's one-stop solutions encompass AIoT modules, AI models, AI agents, global connectivity, and cloud services, driving digital intelligence upgrades in industries such as robotics, consumer electronics, low-altitude economy, intelligent transportation, smart retail, and smart energy. Built on the foundation of seamless software-hardware integration and full-stack coordination, these solutions reinforce Fibocom's position as a leading AIoT platform enabler.AI + Vertical Applications: Accelerating Commercialization and Validation of Smart TerminalsIn recent years, Fibocom has made remarkable progress in the field of smart robotics. Its self-developed embodied intelligent development platform whose name is Fibot integrates AI algorithms, IoT connectivity, and automated control, enabling leading global robotics brands to elevate their product intelligence and performance.A remarkable milestone is the launch of the world's first "vision-only" intelligent lawn mower solution in 2025. Unlike traditional lawn mower solutions, this breakthrough solution requires no physical boundary setup or base station support. Instead, it relies entirely on machine vision and AI algorithms for precise navigation and obstacle avoidance—delivering a true "perception-to-action" experience. "This is more than a technical innovation—it's a glimpse into the future of smart devices," noted Tiger Ying. The solution has already been rolled out across several European markets and earned a 5-star rating from Germany's leading tech media outlet Heimwerker, underscoring Fibocom's global competitiveness in AI innovation.Beyond robotics, Fibocom is redefining the value of 5G FWA through its SkyEngine solution, which combines AI computing with advanced connectivity. Designed to support local AI inference, concurrent multi-tasking, and cross-device collaboration, the platform serves as a powerful "super-intelligent agent" at the edge—enabling applications in smart homes, remote work, and multimedia processing. It also empowers telecom operators to evolve from bandwidth providers into intelligent service enablers.Driven by a dual-engine strategy of "algorithms + computing power," Fibocom is accelerating the commercialization of vertical AI applications—solidifying its transformation from a wireless communication module supplier to a global leader in AI-driven solutions.Addressing AI Integration Challenges with Systematic InnovationAs edge AI adoption accelerates, Fibocom is proactively addressing the associated challenges through a comprehensive and forward-looking approach. In the realm of data security and privacy, the company adheres to a "data minimization" principle from the design stage—enhancing on-device processing to reduce data exposure risks and strengthen user trust.For power-sensitive applications, Fibocom delivers high-performance, energy-efficient AI hardware solutions, optimizing resources from semiconductor architecture to algorithmic efficiency. Meanwhile, to address the complexities of computing scalability and ecosystem integration, Fibocom is building an open, flexible collaboration platform that enables customers and partners to co-create value, share in the AI-driven ecosystem, and accelerate real-world deployment across diverse use cases."We are committed to long-term value creation and scalable commercialization, with a sharp focus on edge AI solutions," said Tiger Ying. "Modules are not only the nearest data entry point to end devices—they are also the critical gateway to unlocking the full potential of AI."At the forefront of the AI revolution, Fibocom continues to be a foundational force in enabling intelligent connectivity, propelling the AIoT industry into a new era of transformation.
Quebec-based startup Femtum, a trailblazer in laser technology, is carving out a critical role in the evolution of semiconductor manufacturing.Spun off from Laval University's Centre of Optics, Photonics, and Lasers in November 2017, the company was founded by Louis-Rafaël Robichaud and Simon Duval, who are respectively the CEO and CTO.Initially, the focus of the company was on the laser sources. Since it was a spinoff, it had more than 20 years of research that Femtum acquired through a transfer of technologies by licensing the IP from the university.Robichaud and other experts on the team bring a wealth of knowledge to Femtum, including laser engines, fiber laser engines, all the assemblies, and the processing steps that are required to make them viable."Still today, Femtum is the only mid-infrared short pulse fiber laser in the industry," he asserted, highlighting the company's unique technological advantage. This distinctiveness is crucial in addressing the intricate challenges of manufacturing next-generation semiconductors for AI and photonics.Femtum's laser solutions are engineered to enhance precision in essential processes such as trimming and cleaning. In the production of advanced components like Co-Packaged Optics (CPO), High Bandwidth Memory (HBM), and silicon photonics, the necessity for meticulous cleaning is paramount."One dust particle can kill a full CPO package," Robichaud warned, emphasizing the potentially devastating impact of contaminants and the critical role of Femtum's cleaning technology in preventing defects and boosting production yield.Robichaud emphasized that where Femtum distinguishes itself from other competitors is the fact that it not only provides the laser source, but also the optical end, how to focus the beam, how to condition the beam, and a fiber beam delivery. "We have fast shutters. We have optics to focus it really well. We also provide a software suite, which is pretty advanced to do the laser cleaning and laser trimming," said Robichaud.Versatile in laser solutionsThe company's technology is not just about cleaning; it's also about enabling new manufacturing processes. "We can really enable new processes. Epoxy is widely used for the bonding of parts together," Robichaud explained, pointing to the versatility of Femtum's laser solutions.Femtum has already demonstrated the efficacy of its solutions through pilot programs with key industry players. Its partnership with ficonTEC, a global leader in fiber-attached systems, has been instrumental in validating the technology at wafer scale."ficonTEC is the leading supplier of fiber attached systems in the world. They're a German company. They're the suppliers of many of the biggest Taiwanese OSATs," Robichaud stated, underscoring the significance of these collaborations.Femtum is also collaborating with other major entities, including INO and C2MI in Canada, to further refine and expand the applications of its laser technology.As Femtum looks to the future, scalability is a key focus. "The aim of Femtum is to achieve a thousand laser solutions per year," Robichaud revealed, outlining the company's ambition to significantly increase its production capacity. This vision includes the construction of a new factory in Quebec City.Seeking strategic investmentsTo fuel its expansion, Femtum is actively pursuing strategic investments. "We are raising Series A as we speak, and we hope to close this round by the end of this year," Robichaud confirmed. The company is particularly interested in attracting Taiwanese investors, recognizing the pivotal role of Taiwan in the semiconductor ecosystem."Taiwan is a key market, if not the top three markets for sure for Femtum," Robichaud emphasized, highlighting the country's importance for the company's growth strategy. He also mentioned that Femtum already has distributors in Taiwan.Femtum's business model is centered on providing comprehensive laser solutions. "We are a Femtum inside play. We co-sell our solution together with the machine integrator," Robichaud explained. This collaborative approach ensures that Femtum's technology is seamlessly integrated into the broader manufacturing process.Looking for investors, Robichaud hopes to find those that are more than just financial backers. "We're not looking only for people that can provide money, but also people that can bring value to the company, open doors and make those strategic partnerships or strategic projects or customers or suppliers, things like that," he said.With its sights set on revolutionizing semiconductor manufacturing, Femtum is poised for significant growth, driven by its cutting-edge technology and strategic vision.
Vancouver-based startup Clirio is transforming how engineering projects are visualized and managed through its cutting-edge platform that integrates augmented reality (AR), LIDAR scanning, subsurface data, and 3D visualization.Founded in 2020 as a spin-off from an applied earth science consulting company, Clirio aims to help engineering projects make better risk-informed decisions regarding large infrastructure projects such as transportation corridors, bridges, tunnels, and mining operations.According to CEO Gerald Magnusson, Clirio was born from the need to address the challenges of traditional engineering practices. "We're trying to help engineering projects make better risk-informed decisions around large infrastructure projects, such as transportation corridors, bridges, tunnels, and tailing dams for mining operations," Magnusson explains.Magnusson further highlights the difficulty for the general public to comprehend complex projects: "As you can imagine, for the layperson, looking at hundreds of 2D drawings is often challenging to understand what the scope and scale of the problem of the project is." Clirio's platform addresses this by "taking that data and putting it into the middle of the room so that everyone can ask questions based on the initial model, so everyone starts learning and growing their knowledge base around the project," says Magnusson.Core technology and unique advantagesClirio's core technology combines augmented reality, LIDAR-based scanning capabilities, subsurface data, and advanced 3D visualization. The platform enables users to capture, visualize, and collaborate on geospatial and project data in immersive environments.What sets Clirio apart, Magnusson states, is "its ability to integrate real-world 3D data, or what we often refer to as digital twins, into highly collaborative cloud-based workflows, allowing those teams to work anywhere in the world with one another". Unlike traditional BIM solutions, Clirio's tools focus on delivering real-world, spatially accurate visualizations.Scalability and data interoperabilityClirio's platform is designed with scalability in mind, capable of supporting organizations from small teams to large enterprises. The cloud-native architecture ensures that data storage, processing, and collaboration grow seamlessly. The platform also utilizes field-gathering tools on smartphones and tablets, enabling users to keep data current throughout a project's lifecycle."We've prioritized data interoperability and offer integrations with most commonly used data formats," Magnusson notes, emphasizing Clirio's ability to seamlessly integrate into clients' existing spatial data workflows. The platform supports various GIS systems, subsurface information, CAD platforms, and data formats, providing a comprehensive view of both above-surface and below-surface data.Customization and diverse applicationsClirio's platform is also highly customizable, allowing users to tailor it to their specific industries and use cases.Magnusson illustrates the platform's versatility: "Large-scale infrastructure engineering and construction projects can utilize geo hazard mapping and annotation tools. Archeologists could use it for digital preservation and virtual tours of excavation sites. Natural resource industries can adapt the platform for pipeline monitoring and environmental impact assessments," and even "Mining operators can collect geo-referenced 3D scans of changing conditions on site to provide safety briefings for workers that are just arriving on-site".Business model and financialsClirio operates on a subscription-based SaaS model, with pricing tiers based on usage volume and organizational size.Small teams can have access to affordable plans with limited user seats and digital workspaces. "Our tiered plans include a free tier where a user has access to one workspace. There's the pro tier for $49 US per month, and then there's a premium tier for $89 US per month". For larger clients, enterprise plans offer unlimited access and custom feature development.Clirio's annual recurring revenue is approximately CA$400,000, with CA$350,000 from services and CA$50,000 from the product side. The company's current burn rate is about CA$42,000 per month.Future roadmap and expansionClirio's future roadmap includes integrating AI to analyze geospatial data and provide actionable recommendations. The company also plans to add more data integrations and enhance its 3D annotation and design tools.Magnusson said it is also focused on expanding its hardware compatibility and establishing strategic partnerships in regions like Taiwan. He highlighted the company's interest in partnering with AI data analytics firms, display hardware manufacturers, and supporting Taiwan's infrastructure development.Looking ahead, Magnusson envisions Clirio continuing to grow its feature set based on user feedback and expanding its reach through joint ventures and strategic partnerships. "Really it is about continuing to grow our feature set based on, on what our users are telling us that is, that is critical for their types of work," Magnusson states. Ultimately, Clirio's vision is to support companies in making better-informed decisions on their projects, ensuring the best outcome for everyone involved.
Letenda, a company with a strong vision to become a major player in the transportation industry, is making significant strides in the electric bus market. With a focus on innovation, sustainability, and collaboration, Letenda aims to transform public transit.Nicolas Letendre, Letenda's CEO and founder, with over 15 years of experience in the transportation vehicle manufacturing industry, including roles at tier-one suppliers and Bombardier Aerospace, identified a crucial gap in the North American bus market. "My family comes from an entrepreneur family, and I always wanted to start a vehicle manufacturing company," Letendre explains. This passion, combined with technical expertise and a keen business sense, led to the inception of Letenda.In 2013, Letendre began investigating the bus industry, engaging with customers and suppliers to understand their needs. A key takeaway from these interactions was the demand for a mid-sized bus, a segment underserved in North America, where the market has been dominated by 30-foot and 40-foot full-size buses since the 1960s. "The customers told us we need smaller buses than the full-size bus that we have in North America," Letendre states, highlighting the need for a 9-meter bus for transit and shuttle services.Innovation at the forefrontLetenda's journey began in 2016, with Letendre participating in various accelerators to refine the business plan and model. Securing initial investments in 2017, Letendre focused on developing a battery-electric bus program that addresses key customer requirements: zero emissions, improved passenger accessibility, enhanced maneuverability, and a good range.Letenda's commitment to innovation is evident in its design and technology. The company's buses feature a distinct shape, improved passenger cabin and accessibility, enhanced driver field of view for increased safety, and energy efficiency, and computer-controlled systems for data provision. A rear steering system enhances maneuverability, allowing drivers to navigate tight corners with ease. "So you drive the bus now like a car and not like a bus anymore," Letendre explains.Targeting the public transit marketLetenda primarily targets the public transit market, including municipal transit, airport shuttles, and university shuttles in the USA. The company also recognizes the emerging demand for on-demand transit services in smaller towns, where smaller buses like Letenda's are ideal. "Like the Uber, but for buses, and you don't need a full-size bus for that service, they take less people, so a smaller bus makes sense," Letendre points out.Strategic manufacturing and expansionLetenda's business model involves collaborating with manufacturing partners to produce its buses. This approach allows for efficient production ramp-up and scalability. Initial manufacturing will occur in Canada to serve launch customers, with plans to utilize a manufacturing partner's US site to comply with "Buy America" requirements for public transit projects in the United States.The company is also exploring opportunities to expand its manufacturing and market reach. Letendre expresses interest in collaborating with Taiwanese manufacturers to serve markets in different regions, leveraging Taiwan's capabilities and Letenda's innovative platform.Competitive and sustainableLetenda is committed to providing competitive solutions in the electric bus market. The company's buses are priced competitively with major competitors like BYD, and Letenda has the advantage of being eligible for FTA (Federal Transit Administration) projects and avoiding surtaxes. "We are competitive with the other electric buses. BYD is a Chinese company, and is banned from applying to FTA like public transit projects", Letendre explains.Letendre also recognizes the importance of government incentives in driving the adoption of electric buses. In Canada and the USA, public transit operators utilize grants to purchase buses, a model that Letenda is familiar with. "And In Canada, we have the incentive for research and development (R&D)," Letendre emphasized that the sound ecosystem of R&D is very helpful for the company.Data monetization and AI integrationLetenda's business model extends beyond bus sales to include aftermarket parts and data monetization. The company plans to leverage the data generated by its buses to provide valuable insights to customers, enhancing their operations and services.Furthermore, Letenda is embracing cutting-edge technologies like AI to enhance its offerings. The company is integrating AI-powered passenger counting devices to provide valuable data to customers.Letenda is currently in a Series A fundraising round to support its commercialization phase. The funds will be used to purchase parts and work with the manufacturing partner to deliver buses. "We are planning to achieve our breakeven point executing the Series A plan," Letendre states.Collaboration and government incentivesLetenda is coming to Taiwan in May for the Computex and actively seeks strategic partnerships to expand its market presence and leverage local expertise. The company has initiated discussions with Taiwanese bus manufacturing companies to explore collaboration opportunities.As an innovative bus company based in Quebec, Letenda is poised to make a significant impact on the transportation industry with its innovative electric bus technology, strategic partnerships, and commitment to sustainability.
In an era dominated by cloud-based AI, a growing movement toward edge computing is emerging. Canadian edge AI startup Picovoice offers a full portfolio of voice AI and LLM technologies, distinguishing itself by delivering cloud-level performance directly on edge devices, combining high performance with privacy and low latency.Founded in January 2018 by Alireza Kenarsari, Picovoice aims to accelerate the transition of AI processing to where data is generated – at the user's fingertips.Kenarsari observed the inherent inefficiencies of cloud-dependent AI assistants, questioning the necessity of sending even simple voice commands to distant data centers for processing."If you tell Alexa to turn off the lights, does it really need to go through half of the planet, go to a data center, and come back? Probably not," Kenarsari said. "I knew what was going to happen, so my goal was to accelerate that transition."He envisioned a future where AI could operate more locally, mirroring human intelligence. This vision, combined with his entrepreneurial spirit, led him to establish Picovoice.Kenarsari's journey to Picovoice is marked by a rich blend of experiences. His career began with stints at three different startups in Vancouver, where he witnessed the full spectrum of outcomes from acquisition and IPO to complete failure. This was followed by a role at Amazon, focusing on machine learning-based financial fraud detection. His close interactions with the Alexa unit at Amazon provided him with key insights that would shape his vision for his startup.End-to-end optimizationPicovoice's core technology is distinguished by its end-to-end optimization. The company owns its data pipelines, training mechanisms, and inference engines, enabling it to fine-tune its AI models for optimal performance on edge devices. This approach allows Picovoice to match the accuracy of cloud-based APIs while providing the low latency and reliability crucial for real-time applications.What differentiates Picovoice's solution from other edge-computing competitors? "Many edge deployment solutions use post-training optimization, where a pre-trained model (often open source) is modified to be smaller and faster. This approach has limitations: the original model isn't designed for edge deployment, restricting potential optimizations," explained Kenarsari. "Additionally, reliance on open-source runtimes like PyTorch and TensorFlow limits the optimization techniques available." These limitations make it difficult to achieve cloud-level accuracy on edge devices. That's why Picovoice addresses this by creating its own data pipeline and training mechanism.To empower non-developers to design voice-enabled product interactions, Picovoice has developed a web-based platform that simplifies the creation and customization of voice commands and wake words. It adopts a customer-centric approach, tailoring its business model to suit the unique needs of each client. As Kenarsari notes, "If you're making an AI PC, your constraints are very different compared to someone who is making a $5 million surgical robot."Its solutions are designed to be cross-platform compatible, supporting a wide array of operating systems and hardware configurations from Linux, macOS, Windows, Android, iOS, Chrome, Safari, Edge, and Firefox to NPU, GPU, CPU, MPU, and MCUs. "This versatility is particularly valuable for enterprises with diverse product portfolios," said Kenarsari. This is especially valuable for large enterprises that want to provide consistent user experiences across different devices.Value proposition for the edgePicovoice targets industries where privacy, reliability, and real-time processing are paramount, including consumer electronics, automotive, healthcare, public safety, and government tech. The company's business model is B2B-focused, offering cost-effective solutions for high-volume applications. Its technology is being utilized in various innovative applications, including LLM-based voice assistants, agentic AI, and even NASA's next-generation spacesuits. In public safety, Picovoice's on-device solutions ensure that sensitive data remains secure.Picovoice initially earned recognition by delivering technically advanced solutions that outpaced competitors, gaining credibility among sophisticated technical buyers. As the company evolves, expanding its product portfolio and refining its go-to-market strategy, its value is increasingly recognized by non-technical stakeholders as well. With a commitment to transparency, Picovoice offers open access to its technology and resources. Its flexible pricing model is designed to adapt to the unique needs and scale of each customer.Looking aheadLooking ahead, Picovoice's vision is to power a billion devices with its AI technology. It is actively expanding its team and exploring strategic acquisitions to further its reach and impact.The company remains committed to continuous innovation, investing heavily in research and development to stay at the forefront of the AI revolution.Picovoice has achieved profitability and secured significant multi-year deals with Fortune 100 companies. While the company is not actively seeking fundraising, it is open to strategic acquisitions to expand its team and capabilities.
In an exclusive interview, Mariia Zhuldybina, CEO and co-founder of TRAQC, a Canadian startup, shared her vision for transforming quality control in additive electronics manufacturing. The company is pioneering the use of terahertz radiation to inspect the functionality of printed materials in real-time.Zhuldybina's journey began during her PhD research in terahertz metamaterials, where she visited industrial facilities producing printable electronics. "I was fascinated about the production line and how it is working at huge volume and how many we can produce in a very short time," Zhuldybina said, but she also noticed that quality control methods were outdated. This realization sparked her desire to bring her research to the industry, developing a solution for in-line, non-contact quality control.TRAQC's technology uses terahertz light to measure the functionality of printed materials, similar to how our eyes see different colors of light. "Our system is actually measuring how the signal of terahertz light is reflected which corresponds to actual functionality of the printed production of the printed traces. We do it in a similar way to your eyes, for example exploring a rainbow: what is red what is blue what is green, and we can also do it with terahertz to explore what is good electrical properties, what is bad electrical properties," Zhuldybina explained.Key AdvantagesTRAQC's system offers significant advantages over traditional quality control methods. It can operate at speeds up to 50 meters per minute with 95% precision. In contrast, traditional camera solutions only inspect the surface and cannot assess functionality, while probe methods are manual and can damage the products. According to Zhuldybina, "sometimes people throw away kilometers of printed devices because there was a lack of quality control during the production in line."Applications and MarketTRAQC's technology targets various additive electronics manufacturing sectors, including sensors, RFIDs, NFC antennas, medical devices, wearable devices, and solar cells. The company is particularly focused on the solar energy sector, where there is a growing demand for fast, cheap, and thin solar panels. Zhuldybina emphasized the urgency, stating, "We need to go from producing like 360 gigawatt of solar panels per year to 3,000 gigawatts. We need to jump almost 10 times within five years."Competitive Edge and PricingWhile there are existing quality control methods like probe methods, camera systems, and profilometers, TRAQC's solution offers a more comprehensive approach. "All the solutions are missing something what we complement," Zhuldybina noted. Although TRAQC's hardware, priced at $75,000, is more expensive than a simple multimeter, it provides added value through consistent data collection, AI integration, and real-time feedback. The company also employs a subscription model for its AI-powered services, costing around $3,000 per month.Installation, Scalability, and Taiwan's Supply ChainsTRAQC's system is designed for easy integration into existing production lines, requiring only about 30 minutes for installation. The company is currently raising $1 million in pre-seed funding to scale up production, deploy pilot systems, and accelerate market entry.Recognizing the importance of strategic partnerships, TRAQC is also looking to leverage Taiwan's robust supply chains. "Yes, we are very open for this opportunity, and not only in printable electronics but in terms of the technology which we are using," said Zhuldybina, indicating a dual interest in both printable electronics and the broader technology supply chain in Taiwan.Vision for the FutureLooking ahead, TRAQC aims to become the standard tool for the printed electronics industry, providing a comprehensive solution for quality control and process improvement. The company is also committed to sustainability, with the goal of reducing waste and energy consumption in manufacturing. According to their estimates, a single TRAQC machine can reduce CO2 emissions by 150,000 kilograms of CO2 equivalent.TRAQC's innovative approach has the potential to significantly impact the additive electronics manufacturing industry by improving efficiency, reducing waste, and enabling the production of high-quality products.TRAQC's system offers significant advantages over traditional quality control methodsPhoto: CompanyTRAQC's system is designed for easy integration into existing production linesPhoto: Company
Procero, a company focused on enabling edge AI use cases, is making significant strides in empowering businesses to run AI applications locally and within their own environments. In a recent interview, Board Director Dr. Richard Reiner discussed Procero's mission, its technology's potential, and its views on the evolving AI landscape.Procero's core focus is to enable AI applications, including Large Language Models (LLMs) and other generative AI, to operate on-premise, addressing critical concerns around data privacy, cost-efficiency, and latency. Dr. Reiner highlighted the increasing demand for edge AI solutions, driven by the need to avoid reliance on large, cloud-based environments."The demand for running AI on the edge is very strong, but the availability of edge AI solutions has been limited," Dr. Reiner stated, emphasizing Procero's commitment to enhancing the capability of edge devices to handle more complex AI use cases with greater performance.Procero sees strong potential across several vertical markets, including advanced manufacturing, cybersecurity, network intelligence, telecom-based edge computing, smart mobility, and healthcare. Dr. Reiner pointed out healthcare as a key area where edge AI solutions are crucial due to privacy regulations.Strategic partnerships across the AI ecosystemProcero is actively seeking to collaborate with key players across the AI technology ecosystem. This includes:*Semiconductor companies: Procero aims to partner with companies producing NPUs, GPUs, or TPUs, particularly those focused on edge devices for AI PCs, AI-enabled mobile devices, and IoT devices. Procero's software solution can significantly expand the capability to run AI applications on these systems.*Device OEMs: Procero also targets device Original Equipment Manufacturers (OEMs) that integrate semiconductor and memory technology into finished products like IoT devices, mobile phones, and laptops. The company offers its software to enhance AI capabilities at the device level.*Application vendors: Recognizing the importance of application-specific AI implementations, Procero seeks to collaborate with vendors developing AI-driven applications such as chatbots, image generation, and video generation for various devices. Procero's edge-optimized inference engine can be embedded within these applications to improve performance.Early success and strong IP moatFounded at the end of 2023, Procero has quickly advanced from its R&D phase to having its first commercially available product and is engaging with major companies. The company was founded within a venture lab in Montreal called Tandem Launch, and its technology is based on intellectual property licensed from leading university research institutions as well as additional IP developed within the company.Procero has a strong IP moat with multiple families of issued patents and ongoing patent applications. Dr. Reiner believes this robust IP protection differentiates Procero and safeguards its innovations.Experienced leadership teamProcero boasts a strong leadership team, including CEO Audrey Larocque, who brings extensive experience in venture-backed companies and fundraising, and CTO Dr. Vahe Vardanyan, a seasoned AI expert with over a decade of experience, including a background at one of Canada's leading government research AI institutions and Ubisoft. The team also includes VP of Engineering François Painchaud, who has led engineering teams at companies like Autodesk, Epic Games, and Unity Technologies.Impact of DeepSeek and AI advancementsThe emergence of models like DeepSeek has brought both excitement and controversy to the AI field. While acknowledging the concerns around the training process of DeepSeek, Dr. Reiner highlighted the novel techniques it introduced that have led to efficiency gains."They did introduce a number of novel techniques that lead to gains in efficiency," Dr. Reiner noted, adding that these techniques are being adopted by various organizations.Dr. Reiner confirmed that Procero's inference engine is compatible with and enhances the performance of models like DeepSeek, demonstrating its adaptability and value in the evolving AI landscape.Business model and future visionProcero's business model centers around licensing its technology to companies who will bundle it with their products, including software solutions, devices, and applications. While licensing terms are currently negotiated on a case-by-case basis, Procero is working to establish replicable formulas.Looking ahead, Procero is focused on expanding its hardware platform coverage, optimizing its technology for new processor designs, and increasing its support for various AI model types. The company is also committed to growing its team with top-tier talent and is open to establishing R&D centers globally.With a clear vision and a strong technological foundation, Procero is well-positioned to capitalize on the growing demand for edge AI solutions and drive innovation in the AI era.Procero inference engine solution for edge AI. Credit: ProceroBenefits of Procero inference engine for edge AI. Credit: ProceroThe architecture of the Procero inference engine for edge AI. Credit: Procero
Smiths Interconnect, a leading provider of innovative solutions for critical semiconductor test applications, today announced the launch of DaVinci Gen V, the latest flagship product in its DaVinci series portfolio.The product rigorously tests semiconductor chips during the manufacturing process to ensure they provide ultra-reliable and repeatable performance. This is important because these chips are used in a range of areas essential to everyday life, including AI accelerators, automotive systems, and next-generation 6G communications networks, which are set for global rollout over the next few years.The DaVinci Gen V solves the fundamental industry challenges of impedance tuning – an essential part of circuit design that enables the maximum transfer of signal power – and signal integrity – the measurement of the quality of an electrical signal as it travels through a circuit. By tackling these challenges the DaVinci Gen V ensures reliable, super-high performance in electronic systems and testing.Brian Mitchell, Vice President of Smiths Interconnect's Semiconductor Test Business Unit, said: "This innovation empowers semiconductor manufacturers to test faster, more reliably, and with greater precision – ensuring their products meet the demands of AI, 6G, and next-generation technologies. More than just a product, DaVinci Gen V embodies Smiths Interconnect's commitment to advancing semiconductor testing through continuous innovation."The technical specificationThe product delivers breakthrough high-speed signal transmission performance. It enables unprecedented digital signaling speeds of up to 224 Gbps PAM4 for AI accelerators and beyond 100 GHz for 6G communications – these speeds are crucial for supporting the growing demand for massive data transfer.The product also supports the growing complexity of modern integrated circuits, accommodating a 40% increase in next-generation ASIC (application-specific integrated circuits) sizes. ASICs are powerful computer chips that combine several different circuits all on one chip, in a 'system on a chip' design.As next-generation integrated circuits continue to evolve – doubling in bandwidth and computational power every two years – DaVinci Gen V is designed for seamless integration. It maintains full compatibility with existing test hardware, allowing manufacturers to transition effortlessly, reduce development cycles, and accelerate time-to-market.Brian Mitchell, Vice President of Smiths Interconnect's Semiconductor Test Business Unit.Credit: Smiths Interconnect
SK hynix (or "the company") announced today that it recorded KRW17.6391 trillion in revenues, KRW7.4405 trillion in operating profit (with an operating margin of 42%), and KRW8.1082 trillion won in net profit (with a net margin of 46%) in the first quarter this year.Both revenues and operating profit are the 2nd highest records following last quarter when the company achieved its best quarterly results. Operating margin improved by 1%p compared to the previous quarter to 42%, resulting in 8th consecutive quarterly growth.SK hynix explained that memory market ramped up faster than expected due to competition to develop AI systems and inventory accumulation demand. The company responded to the demand with an expansion in sales of high value-added products such as 12-layer HBM3E and DDR5.The company believes the strong financial results despite a low seasonality reflect its outstanding competitiveness compared to the past. The company plans to focus on enhancing the business fundamentals to achieve distinguished financial outcome, even in times of market correction.Based on the financial result, cash and cash equivalents increased by 0.2 trillion won to KRW14.3 trillion at the end of the first quarter, compared to the end of 2024, leading to an improvement in the debt and net debt ratio to 29% and 11%, respectively.SK hynix will continue to strengthen collaboration with supply chain partners to meet customer needs despite demand fluctuation amid global uncertainties.Due to the characteristics of the HBM market that supply volume is mutually agreed a year in advance, the company maintains its earlier projection that HBM demand will approximately double compared to the last year. As a result, sales of 12-layer HBM3E are expected to favorably increase to account for over 50% of total HBM3E revenues in the second quarter.In addition, the company started to supply LPCAMM21, high performance memory module for AI PC, to customers in the first quarter and plans to supply SOCAMM2, a low-power DRAM module for AI servers, when demand ramps up.1 Low-Power Compression Attached Memory Module (LPCAMM2): LPDDR5X-based module solution that provides power efficiency and high performance as well as space savings. It has the performance effect of replacing two existing DDR5 SODIMMs with one LPCAMM22 Small Outline Compression Attached Memory Module (SOCAMM): A low-power DRAM-based memory module for AI serverFor NAND, the company plans to actively respond to demand for high-capacity eSSD, while maintaining profitability-first operation with cautious approach for investment."In compliance with the 'Capex Discipline', SK hynix will focus on products with demand feasibility and profitability to enhance investment efficiency," said Kim Woohyun, Chief Financial Officer. "As an AI memory leader, we will strengthen collaboration with partners and carry out technological innovation in efforts to continue profit growth with industry-leading competitiveness."¡½ 1Q25 Financial Results (K-IFRS)*Unit: Billion KRW 1Q25QoQYoY 4Q24Change1Q24ChangeRevenues17,639.119,767.0-11%12,429.642%Operating Profit7,440.58,082.8-8%2,886.0158%Operating Margin42%41%1%p23%19%pNet Income8,108.28,006.51%1,917.0323%¡° Financial information of the earnings is based on K-IFRS¡° Please note that the financial results discussed herein are preliminary and speak only as of April 24, 2025. Readers should not assume that this information remains operative at a later time.
