From identifying promising technologies to creating new business opportunities: How SK hynix Ventures finds the path forward

Источник: SK hynix Newsroom

From identifying promising technologies to creating new business opportunities: How SK hynix Ventures finds the path forward

Source: SK hynix Newsroom

Identifying promising technologies early and investing in them is one challenge. Turning those technologies into businesses is another. That requires assessing what problems a technology can solve, refining it with research and development teams, and evaluating its potential for deployment

•Updated: October 2, 2026

Identifying promising technologies early and investing in them is one challenge. Turning those technologies into businesses is another. That requires assessing what problems a technology can solve, refining it with research and development teams, and evaluating its potential for deployment in manufacturing operations or customer environments.

Over the past decade, SK hynix has gained experience with this process through sustained venture investment. Its corporate venture capital (CVC) activities have evolved from “tech sensing,” which focuses on identifying emerging technologies and companies early, to “path finding,” which seeks ways to connect those technologies to products and businesses. The company has also broadened its collaboration with portfolio companies to include joint equipment development, technology evaluation, and the development of new supply chains and customer opportunities.

▲ A scene from SK hynix Ventures Day, held in Silicon Valley in September

“The biggest lesson we have learned over the past 10 years is that investment can lead to tangible results only when it is backed by the technical judgment of our internal business units and sustained collaboration over time,” said Heejin Chung, SVP and Head of Venture Investment at SK hynix.

SK hynix’s venture investment activities do not end with discovering technologies. Rather than starting with a startup’s technology and then searching for a use case, the company examines the challenges facing its R&D and manufacturing operations and identifies technologies with the potential to address them.

Internal R&D and mass-production organizations participate directly in technology validation. Depending on the results, SK hynix may pursue joint development or equipment deployment and, when necessary, explore ways to connect the technology with customers and markets.

“What we aim to do is turn SK hynix’s R&D and manufacturing infrastructure, as well as its customer touchpoints, into assets that startups can actually leverage,” Chung explained.

Past investments provide a clearer picture of how this approach works in practice.

As semiconductor structures, including HBM, continue to scale down and stacking becomes more complex, conventional inspection methods increasingly face difficulties in detecting defects within stacked structures during the manufacturing process. SK hynix identified FemtoMetrix’s Second Harmonic Generation (SHG) technology as a potential solution to this challenge.

SHG uses light to measure the properties of interfaces* between materials such as silicon and silicon oxide without damaging the wafer. This makes it possible to analyze conditions inside stacked structures through a nondestructive method. SK hynix’s mass-production and R&D teams evaluated the technology by applying it to DRAM and HBM processes. After confirming its applicability, the company moved forward with equipment adoption and has decided to install multiple tools per fab.

* Interface: The boundary where two different materials meet

Infinitesima drew attention as a potential solution for another challenge: measurement-speed constraints. The company developed Rapid Probe Microscope (RPM) technology capable of delivering measurement speeds 10 to 100 times faster than conventional Atomic Force Microscopy (AFM). SK hynix demonstrated the technology’s utility by applying it to pattern monitoring in the DRAM manufacturing processes and has confirmed plans to deploy multiple systems.

What the two cases have in common is that each began with a problem that needed to be solved rather than with the technology itself. After identifying the required technology, SK hynix’s internal teams assessed their applicability to manufacturing processes, and the results then led to equipment purchases.

SK hynix’s collaboration with software company MemVerge followed a somewhat different path.

SK hynix initially focused on technology that pools storage class memory (SCM)* resources across multiple servers, allowing them to be used as memory or storage as needed. At the time, SK hynix saw potential to integrate the technology with a Compute Express Link (CXL)*-based memory pool solution under development.

* Storage class memory (SCM): A next-generation memory tier that combines characteristics of memory and storage * Compute Express Link (CXL): A high-speed interface technology that connects processors, memory, and other system components to expand memory capacity and bandwidth

SK hynix researchers applied MemVerge’s technology in a memory pooling* test environment to evaluate its performance. The two companies later expanded their joint research into management technology that pools memory across multiple servers and allocates it where needed.

* Memory pooling: An approach that combines memory resources from multiple devices or servers into a shared pool and allocates them where needed

As the market evolved, so did the focus of the research agenda. Having studied technologies for efficiently managing memory resources distributed across multiple servers, the two companies are now extending their work into technologies that enable AI systems to store and manage accumulated information over long periods and retrieve it when needed. More recently, they have expanded their collaboration into technologies designed to help AI retain and use the context of previous conversations and tasks over longer periods. They are currently developing an AI long-term memory agent and memory management software framework together, with plans to evaluate their effectiveness in internal services before considering broader applications.

What stands out about the MemVerge case is that the collaboration did not remain tied to the single use case originally envisioned at the time of investment. Instead, SK hynix maintained a long-term relationship with the portfolio company and worked with it to identify new applications for its technology.

Over the past decade, SK hynix has refined its approach to investment and collaboration. Now, the technology landscape in which that approach must be applied is itself changing.

As competition in AI infrastructure expands beyond memory to encompass a broader range of hardware and software as well as the data center as a whole, the role of connecting external technologies and capabilities is also growing. Increasingly, system-level challenges need to be addressed in collaboration with specialized companies across different fields, rather than by developing every technology internally.

In response to these changes, SK hynix is expanding the scope of technologies and opportunities explored by its CVC activities. The company is broadening its investment focus beyond semiconductors and memory-adjacent technologies and into AI infrastructure and related industries, while strengthening co-investment with global venture capital firms, CVCs, and specialized funds. In addition to early-stage startups, SK hynix is also considering growth-stage companies that have demonstrated their technologies and business models in the market.

Held in Silicon Valley in September, the SK hynix Ventures launch event brought the experience and collaboration models built through individual investments over the past decade under a single brand, presenting them through a clearer vision to the global venture ecosystem. The launch marked a turning point, formalizing the operating principles and collaboration methods developed through those investments under the SK hynix Ventures name and giving them a clearer external identity. The goal is to give global startups and investors a consistent understanding of the technologies SK hynix is looking for and how it works with its portfolio companies.

The new brand identity (BI) also reflects this direction. The V-shaped design represents venture companies growing through collaboration with SK hynix, reflecting the brand’s commitment to long-term partnerships.

Today, SK hynix Ventures brings relevant technology and business teams into the process from the investment review stage. After an investment is made, it continues collaboration by connecting portfolio companies with SK hynix’s R&D, manufacturing, and sales organizations. For each company, the parties work together to design the required next steps, from technology evaluation and product development to market expansion.

“We launched SK hynix Ventures to bring our investment experience and future collaboration outcomes together under one brand, with the goal of establishing ourselves as a long-term growth partner,” Chung shared. “Along with expanding the scale of our investments, we want to clearly communicate a consistent set of investment principles and our identity as a partner.”

With FemtoMetrix and Infinitesima, collaboration began with challenges on the manufacturing floor and led to equipment adoption. With MemVerge, SK hynix identified new areas for joint research in response to shifts in technology and market conditions. Moving forward, SK hynix Ventures will expand this approach to investment and collaboration as it explores new technologies and business opportunities across the broader AI infrastructure landscape.

The growth SK hynix Ventures is pursuing goes beyond simply expanding its investment portfolio. At the core of its approach is connecting startup technologies with SK hynix’s R&D, manufacturing infrastructure, and customer touchpoints, working with portfolio companies to shape the next steps — from technology validation and joint development to product application and market expansion.

The search for new opportunities for collaboration continues even after investment, evolving alongside changes in technologies and market conditions. The next step for SK hynix Ventures is to build long-term partnerships by connecting the technologies and assets of both sides, identifying new applications for those technologies, and ultimately translating them into products and business opportunities.

▲ Chung Heejin, SVP and Head of Venture Investment at SK hynix, outlines the vision and roadmap for SK hynix Ventures at SK hynix Ventures Day in Silicon Valley last September.

“We want to be remembered as a partner that believes in founders before the market is convinced, stays with them through execution after investing, and moves at the pace of the founders,” Chung said.

If the past 10 years were about discovering promising technologies and building SK hynix’s approach to investment and collaboration, the next 10 will be about demonstrating how those investments translate into products and business results.

What this article says

Something is unclear? Ask about the article — I will explain in plain words.

Do not want to dig deeper? We will sort it out for you.