Samsung Semiconductor

A career can start with one opportunity. 🚀 For Eva Neill, an internship with Samsung Austin Semiconductor opened the door to a career in environmental management — right in her hometown of Taylor, Texas. After two summers with the Environmental, Health and Safety team, Eva joined SAS as a full-time environmental technician. Today, she’s building hands-on experience while pursuing a bachelor’s degree in Environmental Management. Her journey is a reminder that internships can help students explore new fields, discover their interests and find direction for the future — even when they’re not sure where to start. As Eva puts it, the opportunity has been “a rocket towards my career growth and development.” 👉 Read Eva’s story and discover how an internship helped launch her career at Samsung. https://bit.ly/4ixKCbF #SamsungAustinSemiconductor #WorkforceDevelopment #CommunityImpact #CareerDevelopment

AI infrastructure is evolving faster than ever, and compute is only half the battle. To truly scale, we need to rethink system architecture with memory and storage at the core—solving bottlenecks in vector search, KV caching, composable storage, thermal management, and CXL memory pooling. The OCP Global Summit gave us the perfect stage to showcase how Samsung Memory is tackling the industry's toughest bottlenecks head-on. From SSDs that compute to memory that composes, we are building infrastructure in the open alongside our partners and the OCP community. This is how we scale 'AI Innovations Together.' Watch below to see how Samsung Memory is advancing AI infrastructure for the agentic AI era. https://lnkd.in/eiyB_FpV #SamsungSemiconductor #SamsungMemory #OCPSummit2026 #KVCache #Tokenization

Two campuses. $10.9 billion in economic impact. Supporting Central Texas. Samsung Austin Semiconductor (SAS) contributes to Central Texas through economic activity, job creation, workforce development and community investment. In 2025, SAS’s Austin and Taylor campuses generated $10.9 billion in economic impact across the region, supporting 28,746 jobs and accounting for approximately $2 billion in workers’ salaries. That impact extends beyond employment: • $218.2 million in annual revenue for local taxing districts • $5.9 million invested in workforce development • More than $6 million in charitable giving Together, these figures highlight the economic contributions of Samsung Austin Semiconductor’s Austin and Taylor campuses to Central Texas, spanning employment, local revenue, workforce development and community support. 📄 Read the full story to explore SAS’s 2025 economic impact and community contributions: https://bit.ly/4B8Okk9 🎥 Watch the video to see how these contributions connect with the people and communities of Central Texas through work, learning and community. #SamsungSemiconductor #SamsungAustinSemiconductor #Texas #Austin #Taylor #EconomicImpact #CommunityImpact

Reflecting on a busy few days at the #OCPGlobalSummit! Our Samsung experts took the stage to share how we're scaling AI infrastructure and advancing interoperability through workload-driven memory and storage innovation.   Here's a quick snapshot of the sessions: 💠 Memory for the AI Inference Era – Disaggregated CXL memory architectures, flexible memory pods, and KV$ infrastructure optimized for agentic AI workloads 💠 CXL & Near-Data Processing – Next-gen CXL evolution, extensible near-data processing frameworks, and advanced memory fabric orchestration 💠 Open Ecosystem Collaboration – Aligning frameworks and APIs with industry partners to enable a truly unified cross-vendor AI ecosystem 💠 Enterprise SSD Innovation – Performance testing, I/O prioritization, and hardware integrity in advanced two-phase immersion cooling environments 💠 Datacenter Storage Security – Protecting data at rest and enabling secure datacenter sanitization via OCP L.O.C.K. and FIPS compliance   Our collaboration with OCP and industry partners continues as we co-design open, scalable infrastructure for what comes next.   Thank you to everyone who joined our sessions and conversations!   https://lnkd.in/eiyB_FpV #SamsungSemiconductor #SamsungMemory #KVCache #Tokenization #CXL #SSD

As the world enters the era of Agentic AI, Samsung and AMD have collaborated based on the shared understanding that memory expansion using CXL memory increases CPU-accessible memory capacity and bandwidth, thereby enabling support for a larger number of concurrent users. In particular, as the number of agents increases, the memory required for their smooth operation rises and the memory space needed to store the KV cache they generate also increases linearly. In this evaluation, we utilized Samsung CXL memory supporting the CXL 2.0 interface alongside AMD CPUs and GPUs to verify AI performance improvements as the number of concurrent users increased. We also plan to announce further performance gains achieved in a CXL 3.2 interface environment using CXL 3.2 memory and AMD CPUs in the near future. https://lnkd.in/eiyB_FpV #SamsungSemiconductor #AMD #SamsungMemory #AgenticAI #OCPSummit2026

The exhibit hall doors are officially open at the 2026 OCP Global Summit! We’re thrilled to welcome attendees to booth C41 to experience firsthand how Samsung is scaling the next generation of AI infrastructure. Here is a look at the cutting-edge solutions we are showcasing: Scaling efficiency of KV caching with CMM-D MD310, PM1763, BM1773, and SOCAMM2: 💠CMM-D MD310: Samsung’s CXL 3.2-based 256GB memory expander supporting the PCIe Gen6 x8 interface with blazing bandwidth up to 72GB/s. 💠PM1763: Samsung’s next-generation PCIe Gen6 enterprise SSD designed for AI servers and AI infrastructure.  💠BM1773: A 16-channel PCIe Gen5 NVMe SSD featuring 9th-generation 2Tb QLC V-NAND, delivering massive capacities of up to 256245.76TB with top-tier performance for data-intensive workloads. 💠SOCAMM2: Our next-gen LPDDR-based server memory module, powered by LPDDR5X technology, purpose-built for high-density AI computing environments. Scaling throughput for maximized tokenization with HBM4E, LP5X-PIM, and Groq 3 LPU: 💠HBM4E: Our next-generation high-bandwidth memory solution engineered to support future AI accelerators and large-scale AI infrastructure. 💠LP5X-PIM: Integrates a Processing-in-Memory block into each of the 16 DRAM banks, enabling highly parallel computation and delivering up to 8× higher effective bandwidth than conventional LPDDR5X (as demonstrated in evaluations with Samsung Edge AI accelerator SoCs). 💠Groq 3 LPU: Built on Samsung Foundry’s 4nm process technology for low-latency AI inference, supporting massive AI workloads. Our products and technologies are designed to scale the future of AI infrastructure. If you're at OCP, stop by booth C41—we’d love to connect and show you what's next! https://lnkd.in/eiyB_FpV #SamsungSemiconductor #SamsungMemory #OCPSummit2026 #KVCache #Tokenization #SamsungMAXLab #SamsungGOST

What an amazing amount of people coming out for the #OCPGlobalSummit executive sessions! Just off the stage, Hojun Shim delivered an incredible session: “Scaling Innovation for the AI Era: Keep Your Compute Running at Full Throttle with Highly Scalable Memory.” The numbers shared on stage really put the hardware challenge of the agentic era into perspective: 💠 Context dominates: In the agentic era, it is the context—not the model—driving memory traffic. 💠 Up to 6x Growth: Gigabytes read per generated token have surged roughly 6x in just two years. 💠 Up to 60x Growth: The KV cache created every month has grown 60x, completely outrunning current compression techniques. While capacity has grown across every tier over the same period, it still hasn't kept up with demand. To close this gap, Hojun shared how Samsung is pioneering solutions across three core architectural pillars – Speed, Memory Expansion, and Persistence. An impactful session on what it takes to scale open AI infrastructure for the road ahead. https://lnkd.in/eiyB_FpV #SamsungSemiconductor #SamsungMemory #KVCache #Tokenization #SamsungMAXLab #SamsungGOST

The future of AI and HPC performance depends not only on compute power, but on how efficiently data moves between logic and memory.   As workloads scale, traditional architectures are increasingly constrained by bandwidth, latency, and power challenges. Samsung Foundry is helping address this “memory wall” through advanced logic-memory integration, combining high-bandwidth memory with advanced process technology and packaging innovation to support next-generation AI and HPC architectures.   In our latest Tech Blog, learn how Samsung’s custom HBM can help enable greater bandwidth, lower latency, improved power efficiency, and more flexible system architectures for next-generation AI and HPC systems. Read the full Tech Blog here: https://lnkd.in/g9ungXf8.   Samsung’s custom HBM mockup will also be on display at the OCP Global Summit starting October 13. Come check it out in person, and stay tuned for more.    #SamsungFoundry #HBM #AI #HPC #AdvancedPackaging #Semiconductors #Chiplets #SamsungSemiUS #OCPSummit2026

Scaling Semiconductor Excellence Part 7: From Legacy to Next at Giheung Campus Where does semiconductor innovation go next? At Samsung Electronics’ Giheung Campus, a legacy of semiconductor innovation provides the foundation for the next era. 🏭 From Semiconductor Legacy to Future Innovation • S1 Line: Samsung Electronics’ first production line dedicated to foundry manufacturing. • Semiconductor R&D Center: A research hub dedicated to advancing ultra-fine semiconductor processes. • NRD-K: A next-generation semiconductor R&D complex where semiconductor history meets the future, preparing for the next 50 years. A legacy of innovation. Research that pushes further. A future taking shape. At the campus where Samsung Electronics’ semiconductor history began, the next era of innovation is taking shape. Discover how Giheung Campus is preparing for the future of semiconductor technology. #SamsungSemiconductor #GiheungCampus #ScalingSemiconductorExcellence

AI infrastructure is evolving rapidly, and memory and storage architectures are becoming critical to scalability. Meet Samsung Semiconductor experts at the 2026 OCP Global Summit. 🤝 Join our Executive, Expo Hall, and Breakout Sessions to explore Samsung Semiconductor's approach to addressing AI infrastructure scalability through speed, memory expansion, and persistence. Discover technologies including: 🔹 CXL & Scalable Memory Tiering 🔹 AI Workload & KV Cache Optimization 🔹 Storage Performance & Hardware Integrity Explore how Samsung Semiconductor's memory and storage architectures are helping shape the agentic AI era. 🌐 https://lnkd.in/eiyB_FpV #SamsungSemiconductor #SamsungMemory #OCP2026 #KVCache #Tokenization #SamsungMAXLab #SamsungGOST