First Generation SiMa.ai Edge AI Platform Enters Mass Production Amidst Surge in Company Momentum

SiMa.ai, the machine learning company delivering solutions for the embedded edge, today announced it is experiencing significant company-wide momentum including a rapid rise in customer demand, new innovation accolades, 30% year-over-year employee growth, the introduction of a new partner ecosystem, and new venture funding for a total of $200M raised to date. Importantly, following the initial release of SiMa’s Machine Learning System on a Chip (MLSoC™) silicon, SiMa.ai MLSoC boards and Palette software in January, SiMa.ai is fulfilling its commitment to customers by achieving full characterization and testing for production grade releases of its silicon, boards and software functionality only five months later.

Some estimate that up to 20% of the world’s total power will go to computing by the end of the decade unless new compute paradigms are created. This has significant implications for the $40B+ semiconductor market, which is transitioning from classic compute to 100% machine learning-based over the next decade, because legacy technology has been a bottleneck to innovation. The SiMa purpose built ML platform is needed for any machine that exists between the smartphone and data center – a new form factor where hardware and software are optimized to increase performance and save energy.

SiMa.ai is actively working with more than 50 market-leading companies across manufacturing, retail, automotive, government and aviation. The company is well positioned to meet a rise in customer demand, noting increases in both proof of concept deployments and significant growth in the number of evaluation kits distributed since January 2023. The company’s unique approach is earning widespread validation demonstrating first mover advantage in the form of customer and partner adoption, award wins for its technology innovation, and continued investment from leading venture capital firms:

• In April 2023, SiMa.ai set a new industry standard in embedded edge power efficiency, outperforming industry bellwethers, signaling increased importance of the frames/second/watt measurement paradigm and validating SiMa’s purpose-built architecture. SiMa.ai won the Closed Edge Power category in ML inferencing power efficiency at the MLCommons’ MLPerf benchmark competition.*
• As a testament to their execution, SiMa.ai has increased its total amount of funding raised to date to $200M, which includes participation from new SiMa.ai investors VentureTech Alliance and an individual investment from Navin Chaddha, Managing Director, Mayfield.
• SiMa.ai’s impressive group of investors, leadership team and board of directors, includes former Cadence CEO Lip-Bu Tan and former Xilinx CEO Moshe Gavrielov as Chairman. In November, the company announced that Harald Kroeger, a seasoned automotive executive bringing years of experience working with market leading companies including Bosch, Rivian, Mercedes-Benz and Tesla joined SiMa as a Board Member and President of Automotive.
• This year, SiMa.ai also gained recognition amongst its peers as an environment that fosters talent and individual success. The company was named one of 2023’s best startup employers to work for by Forbes, and has seen its workforce grow by more than 30% in the last year – a true testament to the internal emphasis placed on employee fulfillment and satisfaction.
• In conjunction with entering full production on MLSoC silicon and boards featuring full functionality in the latest Palette software release, today the company formally launched the SiMa Partner Program, which will initially focus on a select group of strategic GTM partners including e-con Systems, Inventec Corporation, LIPS Corporation, and iWave.

SiMa.ai Palette software provides a full set of functionality for developing at the edge, including support for multiple ML models and ML pipelines for functional evaluation, performance testing and tuning. The software production milestone delivers on SiMa.ai’s commitment to bring its full stack edge machine learning system to the masses, meaning complete ML application development at the edge interacting with real-time data, without estimation or projections. The functionality allows customers to use their own proprietary algorithm development and testing, deploying their internal ML models and pipelines to silicon for real-time feedback.

To enable ML deployment at the edge, SiMa.ai MLSoC silicon has achieved major milestones that allow the release for volume production orders, including the complete silicon PVT (process, voltage, temperature) characterization and qualification under multiple JEDEC/ESDA standards. The boards have received Conformité Européenne (CE) and Federal Communications Commission (FCC) certifications for EU and USA with Underwriter Laboratories (UL) certification for safety compliance, undergoing their own testing to assure that the board level specifications are met for volume production.

“The unanimous uptick we are seeing in customer, partner and investor demand continues to demonstrate we are on target with our timing and execution against our founding vision to provide effortless ML for every edge device,” said Founder and CEO Krishna Rangasayee. “The legacy one-size-fits-all chip approach, forcing the same technology powering data centers into ‘smart’ cars, drones, and advanced robotics has become a barrier to innovation. SiMa.ai’s purpose-built MLSoC is ready to unleash the edge.”

The SiMa.ai team is well capitalized to push the boundaries of what is possible with computer vision and machine learning while continuing to extend its lead with accelerated adoption of its innovative MLSoC Platform in the rapidly developing specialized AI chip market.

Industry Validation for SiMa.ai

“SiMa.ai’s any, 10X, pushbutton technical vision aligns with our embedded ‘Oosto Inside’ strategy at and beyond the edge, enhancing existing security and safety use cases and extending into new markets. SiMa.ai excels at performance and power management and we see them as a critical partner to Oosto in helping us to further reduce total cost of ownership and expansion of physical security as a service to businesses large and small,” said Avi Golan, CEO, Oosto.

“A wide array of low-powered devices around the world can use SiMa's chips to add AI functionalities - the applications are endless - and the company's software-centric approach is a core differentiator, lowering the barrier to adoption,” said Kai Tsang, Managing Partner, VentureTech Alliance. “We are thrilled to invest in Krishna's vision of reshaping the edge AI market and the company's deep industry experience and technical expertise made it an obvious choice."

To learn more about how SiMa.ai can help your organization, for more information or to schedule a demo with someone from our team, visit our website at www.sima.ai or drop us a line at [email protected].

About SiMa.ai

SiMa.ai is a Machine Learning company delivering the industry’s first software-centric, purpose-built MLSoC platform. We enable Effortless ML deployment and scaling at the embedded edge by allowing customers to address any computer vision problem while achieving up to 10x better performance at the lowest power. Initially focused on computer vision applications, SiMa.ai is led by technologists and business veterans backed by a set of top investors committed to helping customers bring ML on their platforms.

© Copyright 2023 SiMa Technologies, Inc. SiMa.ai logo and other designated brands included herein are trademarks in the United States and other countries.

*Notices And Disclaimers

SiMa.ai is the first ML hardware startup to submit to the MLPerf Closed Edge Power Division and rank #1 in power efficiency with peer reviewed results. MLCommons’ MLPerf benchmarks measure the performance and power-efficiency of applying a trained machine learning model to streams of data and analyzes the latency, throughput and power efficiency with a methodology that ensures apples to apples comparisons, considered the leading standard of the ML industry for comparing like-for-like systems. In this benchmark, the ResNet-50 vision model was used to inference on data streaming into a SiMa.ai MLSoC platform and accurately measuring the throughput and total power consumed. For additional details visit MLcommons.org.

View source version on businesswire.com: https://www.businesswire.com/news/home/20230613219304/en/