Jorge Myszne

Hear Tristan Cool speak on the panel session: Energy Harvesting - Powering the Future of Sensor Networks and Controls at #SensorsConverge!

4

Sivers Semiconductors and #AyarLabs to expand their partnership on enabling high volume manufacturing of optical I/O solutions for scalable cost-effective AI infrastructure. Read the press release: https://lnkd.in/ds53PdUE #AI #Datacenters #Optical #laserarrays #AyarLabs #Semiconductors #OpticalIO

92

1 Comment

Within the “terahertz technology gap” (100 GHz - 30 THz) lies a vast reserve of bandwidth poised for development. “The carrier frequency of 300 GHz is 150 times larger than two gigahertz where the current cellular network is today,” explains Research Scientist James Greenberg, a program manager at IMRA. “So, take the number of devices we currently have and imagine multiplying that by a factor of 150. That's the biggest driver in going into a higher frequency.” However, the development of user-friendly technologies capable of generating, manipulating, and detecting radiation within these frequency bands is not only technically challenging but also hindered by the limitations of available THz-capable components. At Micro Harmonics, we are developing new components that enable scientists to design better and faster mm-wave and terahertz systems. Read more about how our #mmWave isolators allowed researchers at IMRA to have a significant breakthrough: https://lnkd.in/gtFm-Nk5 #mmWave #MicroHarmonics #rfengineering #IMRA

13

1 Comment

The 5th Accelerating AI Workshop functions as a comprehensive platform, spotlighting advancements in all facets of #AI application development and enabling technologies. Join us on May 8 as we tackle the need for energy-centric edge AI hardware, and delve into topics like at-memory spatial architecture, domain-specific neural processors, complex memory hierarchy, and massive parallelism. https://loom.ly/CNTObs4 Mila - Institut québécois d'intelligence artificielle CIFAR Alberta Machine Intelligence Institute (Amii) SCALE AI | Canada's AI Global Innovation Cluster @digital-global-innovation-cluster NGEN AIoT Canada

2

Mivium's Game-Changing Gallium Nitride Tech Seeks to Outperform Silicon, Reducing Toxicity and Enhancing Efficiency for Data Centers and AI Applications http://s4g.biz/14nov24 tl;dr: Eric Tsai’s Mivium focuses on gallium nitride (GaN) to replace silicon, improving efficiency and sustainability. GaN's fast-switching properties make devices smaller, more powerful, and less energy-intensive. Mivium’s unique, clean GaN production process avoids the toxic materials typical in semiconductors. Nevada’s tech-friendly policies support Mivium’s plans to establish a new “gallium valley” in Reno. By crowdfunding, Eric brings everyday investors into Mivium’s journey, making early-stage investing accessible. https://lnkd.in/e2nunP4g #ImpactCrowdfunding #DiverseFounders #SocialEntrepreneurs #CommunityCapital #ImpactInvestors #RIC #InvestmentCrowdfunding #SuperCrowd #gallium

3

🚀 Join IoT expert Umesh Lokhande for our latest Engineering Hour! Learn how Arm Cortex-M microcontrollers' nested vector interrupt controller (NVIC) simplifies hardware interrupt implementation. 📈 Umesh will explore the interrupt lifecycle, priority levels, and reentrancy interrupts, plus demonstrate managing two external interrupts on an STM32 MCU. Don't miss out on this deep dive into advanced design techniques! 📅 Date: Sunday, August 4th  🕒 Time: 15:00 GMT 👉 Join here: https://lnkd.in/gfGjh_U4 #IoT #Engineering #ArmCortexM #STM32 #EmbeddedSystems #TechTalk #Interrupts #NVIC #Microcontrollers #TechEducation

23

1 Comment

Cerebras Systems, under the leadership of CEO Andrew Feldman, made a significant leap in AI performance with its breakthrough in fast inference. By achieving lightning-speed processing of the world’s largest AI models, the company unlocked new possibilities for real-time, complex applications like multi-agent collaboration and advanced reasoning. This rapid inference technology is set to transform industries, enabling real-time responses and dramatically enhancing the capabilities of emerging AI systems. Feldman emphasized that this progress opens doors to use cases, driving the future of AI beyond simple query-based tasks. Read more here: https://lnkd.in/g__sm6Be

1

As technology rapidly grows more complex, it's crucial to have adaptable hardware that can accelerate software-intensive functions while also adding determinism, redundancy, and power efficiency. Menta SAS eFPGA IP provides 100% flexibility, allowing you to customize the level of adaptability to your precise needs. Fully standard cell compliant across all technology nodes, Menta's eFPGA IP offers the easiest integration and highest reliability, with best-in-class test coverage. #Adaptability #FPGA #eFPGA #Acceleration #EDA #Programmable #AI

9

Meet AGUVIS: A pure vision-based framework for autonomous Computer/Mobile-use GUI agents, operating seamlessly across web, desktop, and mobile platforms without UI code. Key Features & Contributions - Pure Vision Framework: First fully autonomous pure vision GUI agent capable of performing tasks independently without relying on closed-source models - Cross-Platform Unification: Unified action space and plugin system that works consistently across different GUI environments - Comprehensive Dataset: Large-scale dataset of GUI agent trajectories with multimodal grounding and reasoning - Two-Stage Training: Novel training pipeline focusing on GUI grounding followed by planning and reasoning - Inner Monologue: Explicit planning and reasoning capabilities integrated into the model training - Generalization to OOD (Out-of-Distribution): AGUVIS demonstrates impressive generalization on OSWorld benchmark even without desktop trajectory data during training. Project Page: https://lnkd.in/guzkwpbB Paper: https://lnkd.in/gs7wXyJS GitHub: https://lnkd.in/gWeSkFmK Collaboratiion among Yiheng Xu Zekun Wang Junli Wang Dunjie Lu Tianbao Xie Amrita Saha Doyen Sahoo Tao Yu Caiming Xiong

155

6 Comments

One of the ways our engineers stay up to date on trends impacting the RF ecosystem is by experimenting with open-source RF modeling and analysis tools. Here, the team uses the Phyton Scipy.signal Toolbox to explore the theory behind poles and zeros and shares their findings. White Paper | RF Tools to Understand How Poles and Zeros Relate to Q Factor https://bit.ly/3xQQSrB

19

🔦 How fast is fast? 🌟 Let's talk about the speed of light. At 299,792,458 meters per second, light zips through fiber optics at an unimaginable pace. It can circle the Earth 7.5 times in just one second! 🚀 In the world of deep tech innovation, understanding such speeds is crucial. Whether it's fiber-optic communication or the quest for quantum computing, the principles behind light's speed drive our advancements. 🌐✨ The speed of light isn't just a number; it's a reminder of the boundaries we're constantly pushing in science and technology. Let’s keep pushing! 🌍🔬 #TechInnovation #QuantumLeap #FutureTech #PhysicsFun #DeepTech

34

4 Comments

I'm thrilled to share that my review paper on optimization techniques in Network-on-Chip (NoC) architecture has been published in IEEE Xplore! 🎉 This paper reviews recent advances in network-on-chip (NoC) optimization techniques, focusing on addressing challenges related to fault tolerance and congestion management in multicore systems. It explores innovative strategies that enhance the resilience and efficiency of NoC systems, including fault-tolerant routing algorithms, wireless NoC congestion control mechanisms, dynamic voltage scaling techniques, and design methodologies for heterogeneous architectures. The effectiveness of these strategies is evaluated in terms of network latency, energy efficiency, and system reliability. This comprehensive overview provides valuable insights into the current state of the art and future directions in NoC optimization, serving as a resource for researchers and practitioners in the field. Check out the full paper here. https://lnkd.in/dj7UEDBn #NoC #NetworkOnChip #Optimization #FaultTolerance #CongestionManagement #MulticoreSystems #IEEE #Research

12

2 Comments

The future of hardware is here, and QCon San Francisco 2024 (November 18-22) is your gateway to understanding the latest advancements. The Hardware Architectures You Need to Know track, hosted by Yao Yue 岳峣, co-founder of IOP Systems, delves into the evolving landscape of hardware architectures that are set to revolutionize how we build and deploy software. As semiconductors evolve, new architectural changes are emerging to meet the demands of modern workloads. This track offers insights into how software developers can adapt to this increasingly integrated world of hardware and software. Here are the sessions you can’t miss: 👉 Maximizing Deep Learning Performance: Hardware and Software Innovations for Optimizing AI Workloads by Bibek Bhattarai, AI Technical Lead @Intel 👉 Optimizing Custom Workloads with RISC-V by Ludovic Henry, Member of Technical Staff @Rivos 👉 Evaluating and Deploying State-of-the-Art Hardware to Meet the Challenges of Modern Workloads by Rebecca Weekly, VP of Infrastructure @GEICO 👉 High-Resolution Platform Observability by Brian Martin, Co-founder and Software Engineer @IOP Systems 👉 Unleashing Llama's Potential: CPU-Based Fine-Tuning by Anil Rajput, AMD Fellow, Software System Design Eng. Java Committee Chair @SPEC, Architected Industry Standard Benchmarks and Authored Best Practices Guides for Platform Engineering and Cloud and Dr. Rema Hariharan, Principal Engineer @AMD, Seasoned Performance Engineer With a Base in Quantitative Sciences and a Penchant for Root-Causing 🔗 Explore the track: https://bit.ly/3ZOFsjr Join us to stay ahead in the evolving world of hardware and software integration. Book before October 29th to save with our early bird pricing. #QConSF #SoftwareDevelopment #DevCommunity #DeveloperConference

18

Understanding WSE and GPU limitations

4

Interesting to see Silicon Labs claiming their position in the #ambientIoT ecosystem. Watch out for my forthcoming interview with Dhiraj Sogani their senior Director of Product Marketing for more on what they are doing and why. https://lnkd.in/gpcRxCdE

10

Thanks to Paul Gillin and SiliconANGLE & theCUBE for well crafted coverage of Wiliot's first public foray into #genAI. This is a breakthrough in terms of being able to interrogate with natural language the ground truth about the physical world at scale. With AI orchestrating the sources of truth about supply chains in real-time. #ambientIoT provides the data, ground truth and value unlock that genAI needs to go from a land rush to sustainable growth. #AI brings to ambientIoT the ability to process a tsunami of data that enables omniscient retail that doesn't overwhelm the human brain. https://lnkd.in/g8xhFfEw

20

1 Comment

Edge computing not required "A well-balanced GPU fleet became more of a priority than allocating nearby GPUs. For ChatGPT, the biggest bottleneck in response time is how quickly GPUs can stream out one token at a time. This means that GPUs geographically closer to end users is less of a priority because the round-trip time a network request takes matters less, than GPUs being “ready to go.” My career has taught me the importance of computing at the edge for latency reasons. This ended up being less relevant for these workloads, especially when GPUs are so constrained." - Evan Morikawa, Applied engineering team, OpenAI ChatGPT Does this mean edge computing is not valuable for AI? No It means edge computing is not valuable for ChatGTP. When there is no deep understanding, there are sweeping statements. Edge is needed, edge is not needed. Is cloud good? Is cloud bad? Unfortunately, the answer is always - "it depends". In telecom we need a more deep understanding of problem domains outside our industry, to understand where and how we can help. I highly recommend George Orosz's newsletter https://lnkd.in/en8BgKb8

22

3 Comments