SAN FRANCISCO, June 02, 2026 (GLOBE NEWSWIRE) — Ainekko, a startup pioneering open, software-defined AI infrastructure, today announced that the CORE-ET Silicon Platform (ETSP) has been accepted as a project within the OpenHW Foundation, an initiative dedicated to developing and standardizing open source semiconductor intellectual property. As a result, the industry is one step closer to realizing the creation of an advanced ecosystem of open AI silicon solutions.
The CORE-ET Silicon Platform is a collection of open hardware and software building blocks designed for fast, energy-efficient AI inference at the edge. By combining many-core RISC-V compute, MRAM-based intelligent memory, and an open development platform, ETSP provides a foundation for building next-generation embedded AI systems.
“AI infrastructure is going through the same transition we saw with operating systems and cloud platforms,” said Roman Shaposhnik, co-founder and CTO of Ainekko. “Closed systems slow innovation. Open platforms accelerate it. By bringing the CORE-ET Silicon Platform into the Eclipse Foundation’s OpenHW ecosystem, we’re creating a neutral foundation where developers, companies, and researchers can collaborate on the next generation of AI silicon for the edge.”
As AI workloads move rapidly from centralized data centers into factories, robotics systems, drones, and intelligent devices, the industry is searching for hardware platforms that combine performance, energy efficiency, and flexibility. Open silicon initiatives are increasingly seen as a way to accelerate innovation by allowing developers to experiment directly with architecture, tooling, and system design.
An Open Platform for Edge AI
As AI workloads move from centralized cloud infrastructure into factories, robotics, drones, and intelligent devices, developers need hardware platforms optimized for power efficiency and flexibility. CORE-ETSP addresses this need with an open platform spanning silicon architecture, system design, and developer tooling.
The platform includes:
- A many-core 64-bit RISC-V processor architecture with vector/SIMD extensions optimized for parallel inference workloads
- A scalable network-on-chip (NoC) architecture with integrated MRAM-based memory
- Flexible system interfaces enabling deployment as either an accelerator or a standalone embedded platform
- A full-featured architecture emulator enabling rapid experimentation and design iteration
- A development platform and tools for porting and running AI inference frameworks
Enabling Open Co-Design of AI Hardware and Software
CORE-ETSP enables a new model of hardware innovation in which silicon architecture evolves alongside the AI software it runs. By making the architecture, RTL, and tooling openly available, developers can experiment with hardware configurations, optimize AI workloads, and contribute improvements back to the ecosystem.
“Open collaboration is essential to advancing the next generation of semiconductor innovation,” said Florian Wolhrab, Head of OpenHW Foundation. “The CORE-ET Silicon Platform brings together open hardware design, development tools, and AI inference capabilities in a way that aligns closely with the OpenHW Foundation’s mission to enable open, high-quality processor and system IP for the global community. As the whole design is being taped out on a 16nm node, it is transforming into real silicon quickly.”
Governance and Licensing
CORE-ETSP will be developed under the OpenHW Foundation, part of the Eclipse Foundation, which provides neutral governance and strong IP protections for open hardware collaboration. The project will be released under the Solderpad Hardware License v2.1, a permissive open hardware license widely used for semiconductor intellectual property, building on the well-known Apache 2.0 license.
Expanding the Open Silicon Ecosystem
The CORE-ET Silicon Platform builds on Ainekko’s broader effort to make AI infrastructure open and developer-driven. The company previously launched AI Foundry, an open community platform hosting silicon designs, development tools, and collaboration resources for AI hardware innovation. Together, these initiatives aim to create an open ecosystem where developers can build and deploy AI systems without relying on proprietary hardware stacks.
Ainekko and the OpenHW Foundation will both be attending RISC-V Summit Europe 2026, taking place in Bologna, Italy, from June 8–12, 2026. Ainekko CEO and co-founder Tanya Dadasheva will also present a session titled “Developing an Open Agentic SoC,” focused on the company’s open, AI-native approach to programmable silicon design and hardware/software co-development.
About Ainekko
Ainekko is a startup building the first open, software-defined platform for AI infrastructure—extending open-source principles all the way down to silicon. By offering flexible, modular hardware and tooling, Ainekko empowers developers to build efficient, composable AI systems for edge and data center applications alike. Founded by industry veterans with deep expertise in open systems and silicon design, Ainekko is on a mission to make AI hardware accessible, developer-driven, and community-powered. More at www.ainekko.io
About The OpenHW Foundation
The OpenHW Foundation, an industry collaboration hosted at the Eclipse Foundation, is a global non-profit organisation focused on developing and promoting high-quality open source hardware IP, tools, and infrastructure based on the RISC-V architecture. Through industry collaboration and ecosystem partnerships, OpenHW advances open silicon adoption across AI, automotive, cloud, industrial, IoT, mobile, and high-performance computing markets.
Media Contacts: For Ainekko: ainekko@orogroup.com
