Curated by THEOUTPOST
On Fri, 24 Jan, 12:03 AM UTC
2 Sources
[1]
Baya Systems raises $36M to transform AI chiplet design
System-on-chip design startup Baya Systems Inc. said today it has closed on $36 million fuding to support its operational growth and accelerate the development of its software portfolio to meet the needs of the emerging "chiplet" economy. Today's Series B round was led by Maverick Silicon and Synopsys Inc., and saw participation from existing investors like Matrix Partners and Intel Capital. So-called chiplets, or SoCs, have become increasingly popular with the emergence of artificial intelligence, which requires extremely efficient data movement and compute density. SoCs combine various semiconductor components, including central processing units, graphics processing units, neural network accelerators and other kinds of chips to create more efficient computing platforms. However, one of the challenges with putting so many different components onto a single piece of silicon is the need for an efficient design that ensures rapid data movement among the various moving parts. Chipmakers get around this by integrating network-on-chip technology with their SoCs. NOCs are specialized components that can move data around on the chip while minimizing the resources needed to connect them all together. They extend across SoCs, but the challenge lies with multi-die implementations -- where companies want to connect large clusters of SoCs to run more powerful applications and workloads. Simply put, NoCs don't work so well when they're required to communicate with other NoCs. Baya Systems says it has developed a "revolutionary" chiplet-optimized NoC and physical link or PHY interconnect solution that gets around this challenge. Though it's possible to connect NoCs on individual chiplets via PHY today, doing so is inefficient, the company says. That's why it has come up with a new approach, which involves integrating various NoCs into a "unified fabric" that functions like a single, intelligent, global NoC. According to Baya Systems, its technology helps reduce the latency and increase the bandwidth and throughput of NoC-to-NoC communications, while providing greater flexibility in terms of traffic routing in multi-die designs. Baya Systems founder and Chief Executive Dr. Sailesh Kumar said designers of complex SoCs and chiplets that combine CPUs, GPUs and other accelerators would typically rely on "brute-force" solutions, and claims that's unsustainable. "There are too many risks, such as high re-engineering costs, difficulty in scaling, and potentially hitting the market with sub-par metrics," he said. "Baya's performance-focused, software-based approach, coupled with our unique transport and modular fabric IP, is designed from the ground up to produce complex multi-die solutions that are correct by construction with a simplified design process." Baya Systems says its technology optimizes and simplifies chiplet communications, freeing systems designers to work on the value-added part of their systems. In other words, chip designers can instead focus on making their chiplets even more powerful. Maverick Silicon Managing Director Andrew Homan said the challenge of AI is not about providing computing power, but facilitating seamless data movement. "The team at Baya Systems is uniquely positioned to fill this critical gap in the industry with WeaverPro, WeaveIP and its other solutions," he said. Baya Systems says early adopters of its software include the AI chip startup Tenstorrent Inc., which recently raised $693 million in late-stage funding to fuel its ambitions of rivaling Nvidia Corp. It's designing more affordable AI accelerators based on a RISC-V architecture that eliminate the need for expensive components such as high-memory bandwidth. The startup said it plans to announce more partnerships with chipmakers later this year.
[2]
Baya Systems raises $36M for AI and chiplet innovation
Baya Systems, a chip technology company that wants accelerate intelligent computing, raised $36 million in funding. Maverick Silicon led the round, which also included strategic investment from Synopsys and current investors including Matrix Partners and Intel Capital reinvesting into the company. The funding will support operational growth, accelerating the development and deployment of the company's software-driven system IP technology portfolio for system-on-chip (SoC) designs and the emerging chiplet economy. As intelligent computing continues to grow, its demands for AI capabilities, more efficient data movement and compute density are driving the evolution of SoCs into "system-of-chips" models. By using chiplets, this new approach offers scalable performance, optimized power and reduced costs without relying solely on traditional approaches, where further advances have been slow in coming and are growing more expensive. Baya delivers modular solutions designed to adapt to changing needs and take advantage of these benefits for next-gen designs for AI, automotive and data center infrastructure -- while also complying with emerging standards like Ultra Accelerator Link (UALink) die-to-die interconnect for further accelerated AI scaling. "Generative AI and multimodal compute have clearly shown that the real challenge has transitioned from compute engines to data movement and connectivity to truly deliver on the performance and efficiency needs of AI acceleration and scale compute infrastructure and communications," said Andrew Homan, managing director at Maverick Silicon, in a statement. "The team at Baya Systems is uniquely positioned to fill this critical gap in the industry with its WeaverPro, WeaveIP and other solutions." To better support AI and other compute-heavy and data-intensive applications, Baya offers a holistic approach to design, analyze and build complex, highly performant multi-die systems that overcome traditional semiconductors' bottlenecks of data movement and scalability. Baya's foundational WeaverPro software enables continuous refinement of data-driven architecture and micro-architecture development from initial specification through post-silicon tuning, with built-in simulation and workload analysis that ensures the design delivers on KPIs. The comprehensive WeaveIP advanced system IP portfolio, with its unique transport, supports custom and standard protocols, maximizes performance and throughput while minimizing latency, silicon footprint and power, rapidly delivering complex solutions. "Baya Systems has executed ahead of schedule on building the team, the technology and the products that deliver on its vision to solve the high-performance system design challenge for the semiconductor industry," said Stan Reiss, general partner at Matrix Partners, in a statement. "This has uncovered a much larger scope for the company, and in our view, this new infusion of capital is necessary to extend leadership and capitalize on that opportunity." With a high-power team of entrepreneurial leaders and engineers from companies like Apple, AMD, Arm, Intel, Qualcomm, and legendary microprocessor architect Jim Keller as chairman, Baya has rapidly emerged out of stealth, delivered its flagship product to market and is now positioned to grow its market share. "Designing increasingly complex combinations of CPUs, GPUs, neural network accelerators and other processors is a brute-force solution that the industry cannot rely on forever. It simply comes with too many risks: high re-engineering costs, difficulty scaling and potentially hitting the market with sub-par metrics," said Sailesh Kumar, founder and CEO, Baya Systems. "Baya's performance-focused, software-based approach, coupled with our unique transport and modular fabric IP, is designed from the ground up to produce complex multi-die solutions that are correct by construction with a simplified design process." Early Baya customers and partners include Tenstorrent, which has licensed Baya technology for its AI and RISC-V chiplet solutions, and some yet to be announced partnerships, signaling Baya's growing traction and global reach. Baya Systems' tech is chiplet ready, meaning it can be built into a system where a bunch of chips sit atop a silicon substrate in a solution that maximize processing power, energy efficiency and networking speed. The company said it has grown in a targeted, judicious way. Nine months after inception, Baya completed its first license and product delivery. Baya's growth accelerated after it emerged from stealth: the company secured multiple customers and partnerships. The company is close to 50 employees, and its goal is to quadruple bookings and revenue by the end of 2025. As for the inspiration, Baya Systems started with a vision to enable semiconductor system design to efficiently transition to the chiplet era and match the exponentially increasing demands of intelligent compute (the tight integration of different compute elements such as CPUs, GPUs and accelerators that enable the diverse computational requirements of AI). Origins The cofounders -- Sailesh Kumar, Eric Norige and Joji Philip -- previously worked together at NetSpeed and later at Intel following its acquisition. At Intel, they particularly focused on the Xeon family of processors, creating a SoC builder and other tools to address then-present challenges of scaling computational capabilities. Baya Systems was incorporated in March 2023 and came out of stealth in June 2024. Baya is headquartered in Santa Clara, California. With their unique perspective, they recognized that communication bandwidth between compute, memory, and input/output (IO) was becoming a critical bottleneck to the advancement of AI and other data-intensive compute functions, one that existing system architectures could not effectively address. Adding Nandan Nayampally as our CCO allowed the company to think bigger. Nayampally's background at Arm involved helped drive the leadership of Arm's iconic Cortex CPU product line beyond the traditional mobile phone, consumer, and microcontroller worlds. But he also championed Arm's coherent interconnect product lines, which are now the foundation for the data center and automotive businesses. Like the founding team, he clearly saw the need for the emerging AI scale needing specialized high performance solutions. Modern architectures require massively parallel and high-bandwidth data movement, as well as low-latency coherent communication within large CPU clusters. The Baya Systems team is determined to enable seamless and efficient communication within computing systems to unlock their full potential, the company said. The company has a legendary chip design executive as its chairman in Keller. He's technologist with vision, and he shares the belief that the foundation of scale for the AI era is not just about specialized compute elements, but also taking a big-picture view of any given system and enabling hyper-efficient data movement. The cofounders originally met Jim during their overlapping time at Intel. Keller also played pivotal roles in companies such as Apple, Tesla, AMD and a variety of chip startups. A phone call between Keller, Tanuja Rao and Kumar set everything in motion.  Keller shared his unwavering belief in the unique skillset of the core team as uniquely capable of solving this massive problem. Within a short span, the firm managed to secure investment, build a world-class team, acquire our first customer, and deliver a high-performance and unique solution. "[Keller's] keen eye for innovation and understanding of market developments have been and will continue to be crucial to us as a company," the company said. As for competition, Arteris is the current incumbent in the pure-play on-chip SoC fabric arena, and companies like Arm and some EDA vendors have products which compete with parts of Baya's portfolio. However, Baya Systems has a unique focus on building chiplet-ready fabric from the ground up that unifies transactions from different protocols over the same transport. Baya's solutions complement Arm and EDA vendor offerings as well as existing EDA tool flows. Some of the large semiconductor leaders have built technologies in-house that are similar to our fabric platform, but these technologies are customized to their system architectures and are not productized for broader market use. Baya view these companies as potential partners, not competitors, because the Baya platform offers a more cost-effective, faster time-to-market solution suited for a number of custom or ASIC projects. The company seeks to partner with organizations that share its philosophy of breaking down the barriers to developing innovative, highly scalable and reliable SoC and chiplet systems. Current partners include Tenstorrent, Intel, Blue Cheetah, the Open Compute Project and UCle (Universal Chiplet Interconnect Express), and the firm actively contributes to and participate in their ecosystems. In the same vein, Baya is also growing its own ecosystem with partners like processor vendors for CPUs, GPUs and NPUs; EDA companies; DHD and PHY vendors; and foundries and design service houses, all of whom can benefit from Baya's portfolio.
Share
Share
Copy Link
Baya Systems raises $36 million in Series B funding to advance its innovative chiplet-optimized Network-on-Chip (NoC) technology, aiming to transform AI chip design and improve data movement efficiency in complex multi-die systems.
Baya Systems, a system-on-chip (SoC) design startup, has successfully closed a $36 million Series B funding round led by Maverick Silicon and Synopsys Inc., with participation from existing investors Matrix Partners and Intel Capital 12. This significant investment will support the company's operational growth and accelerate the development of its software portfolio, aimed at meeting the needs of the emerging "chiplet" economy.
Baya Systems has developed a revolutionary chiplet-optimized Network-on-Chip (NoC) and physical link (PHY) interconnect solution to address the challenges in multi-die implementations 1. The company's technology integrates various NoCs into a "unified fabric" that functions like a single, intelligent, global NoC. This approach helps reduce latency, increase bandwidth and throughput of NoC-to-NoC communications, and provides greater flexibility in traffic routing for multi-die designs.
As intelligent computing continues to grow, the demand for AI capabilities, more efficient data movement, and compute density is driving the evolution of SoCs into "system-of-chips" models 2. Andrew Homan, Managing Director at Maverick Silicon, emphasized that the real challenge in AI has shifted from compute engines to data movement and connectivity 12. Baya Systems' solutions, including WeaverPro and WeaveIP, are uniquely positioned to fill this critical gap in the industry.
Baya's foundational WeaverPro software enables continuous refinement of data-driven architecture and micro-architecture development from initial specification through post-silicon tuning 2. The WeaveIP advanced system IP portfolio supports custom and standard protocols, maximizing performance and throughput while minimizing latency, silicon footprint, and power 2.
Early adopters of Baya Systems' technology include AI chip startup Tenstorrent Inc., which recently raised $693 million in late-stage funding 12. Tenstorrent is designing more affordable AI accelerators based on a RISC-V architecture, eliminating the need for expensive components such as high-memory bandwidth 1.
Founded by Dr. Sailesh Kumar, Eric Norige, and Joji Philip, Baya Systems boasts a high-power team of entrepreneurial leaders and engineers from companies like Apple, AMD, Arm, Intel, and Qualcomm 2. The company's vision is to enable semiconductor system design to efficiently transition to the chiplet era and match the exponentially increasing demands of intelligent compute 2.
With this new funding, Baya Systems aims to grow its market share and extend its leadership in the industry. The company plans to announce more partnerships with chipmakers later this year and has set an ambitious goal to quadruple bookings and revenue by the end of 2025 12. As the demand for efficient AI chip designs continues to grow, Baya Systems is well-positioned to play a crucial role in shaping the future of AI hardware.
Reference
[1]
[2]
Enfabrica, an AI networking startup, secures $115 million in Series C funding and introduces the ACF SuperNIC chip, promising to revolutionize GPU networking for AI applications with unprecedented performance and scalability.
3 Sources
3 Sources
Ayar Labs secures $155 million in Series D funding from major chipmakers and investors to scale up its light-based chip-to-chip communication technology, promising to revolutionize AI infrastructure.
6 Sources
6 Sources
Databricks raises $10 billion at a $62 billion valuation, highlighting the continued surge in AI investments. The news comes alongside other significant AI funding rounds and technological advancements in the industry.
3 Sources
3 Sources
Celestial AI, a Silicon Valley startup, has raised $250 million in a Series-C1 funding round, bringing its total funding to $515 million. The company is developing photonic fabric technology to enhance AI chip connectivity and efficiency.
5 Sources
5 Sources
EnCharge AI, a startup developing energy-efficient analog AI chips, has raised over $100 million in Series B funding. The company claims its technology can rival desktop GPUs while using a fraction of the power, potentially revolutionizing AI processing at the edge.
4 Sources
4 Sources
The Outpost is a comprehensive collection of curated artificial intelligence software tools that cater to the needs of small business owners, bloggers, artists, musicians, entrepreneurs, marketers, writers, and researchers.
© 2025 TheOutpost.AI All rights reserved