OpenAI’s Jalapeño is a custom accelerator designed specifically for AI inference, particularly the low-latency, multi-chip workloads behind interactive and agentic systems. Its central idea is that useful performance cannot be reduced to peak floating-point operations or memory bandwidth. What matters is how quickly … Read More
2026 Ceva Edge Symposium: Building Intelligence Beyond the Cloud
The 2026 Ceva Edge Symposium will bring together engineers, technology leaders, ecosystem partners, and innovators to explore a major shift in computing: moving artificial intelligence from distant cloud infrastructure into the devices that interact directly with the physical world. Scheduled for November 2 in Hsinchu,… Read More
HBM’s Second Act: When Memory Starts Thinking
Samsung’s Hot Chips 2026 presentation, “HBM Base Die: How HBM Will Evolve Using Advanced Logic Processes,” argues that the base die beneath a High Bandwidth Memory stack is about to become far more than a communications layer. Traditionally, an HBM base die mainly connects stacked DRAM dies to a GPU, TPU, or other processor and … Read More
NVIDIA Vera: Rebuilding the CPU for Agentic AI
Source: Jonathon Evans and Polychronis Xekalakis, “NVIDIA Vera CPU,” Hot Chips 2026. Performance figures are NVIDIA claims, and several results were based on preproduction or unofficial testing.
NVIDIA’s Vera CPU is a server processor designed specifically for agentic AI, where language models repeatedly observe, reason,… Read More
ASML and TSMC’s 12-Inch Photomask Initiative: Technical Significance
ASML and TSMC’s proposed shift from 6-inch to 12-inch photomasks is an attempt to redesign a critical but often overlooked part of advanced semiconductor manufacturing. Photomasks are precision plates carrying the circuit patterns projected by a lithography scanner onto a silicon wafer. In extreme-ultraviolet lithography,… Read More
Semicon West: Transforming Tomorrow One Chip at a Time
SEMICON West 2026 arrives at a pivotal moment for the global chip industry. Scheduled for October 13–15 at San Francisco’s Moscone Center, the event returns to the Bay Area after its 2025 edition in Phoenix. Organized by SEMI under the theme “Transform Tomorrow,” it will bring together semiconductor manufacturers, equipment… Read More
HBM’s Vertical Ascent: Why Packaging Is Becoming the Heart of AI Memory
SK hynix’s Hot Chips 2026 presentation explains how High Bandwidth Memory (HBM) is evolving to meet the extraordinary memory requirements of artificial intelligence. Its central message is that future progress will depend not only on better DRAM circuits but also on advanced packaging, denser vertical integration, improved… Read More
DAC 2026: A Discussion of Who Owns the Intelligence Behind Tomorrow’s Chips and How to Make it Better
The industry is moving from simple documentation chatbots to agentic workflows capable of performing complex, multi-step tasks like design creation, verification, and library characterization. These agents are intended to enhance engineering productivity by roughly 50%. A panel of industry experts met at DAC to discuss… Read More
GenAlpha Brings a Practical Approach for Analog AI to DAC 2026
Agentic AI was everywhere at DAC 2026. Various focus areas were presented that covered many parts of the chip design flow. Verification and debug were a popular topic since so much time is spent there. Analog design is another pressure point. Every chip design has analog content. While the size of the analog blocks may not be huge,… Read More
Crescent Island: Turning Memory Capacity into Agentic AI Throughput
Intel’s Crescent Island is a discrete data-center GPU designed for agentic AI inference, where performance is constrained by more than matrix arithmetic. Multi-step agents impose latency through repeated model calls and tool interactions, consume large memory capacity for weights and long-context key-value caches, and… Read More


Comparing Advanced Packaging from TSMC, Intel Foundry, and Samsung Foundry