Research / Topics
Fundamentals
Primers that assume no engineering degree. How chips, networks and optics work, and why the trade-offs matter.
22 posts
No posts match.
2026
- AI With No Decode (Jev), and What It Means for AI Infrastructure
Jev from TypeSafe answers in one pass instead of writing text token by token. Why LLMs are hard to use inside software, how Jev fixes it, and what a model with no decode phase means for AI Infra
- Oligopoly Equilibrium
Why most semiconductor markets settle at two to three players
- NPO State of the Union
When NPO? Who NPO? Where NPO? Why NPO? Winners, losers, and more.
- Build Your Own DRAM Supply Model
Wafer starts, die size, yield, HBM mix, node migration impacts, etc. Simple enough to follow, but detailed enough to learn from.
- High Bandwidth Flash: The Full Report
Why HBF for inference decode fits, what it does to the DRAM and NAND markets, a worked cost model, who moves first, and risks
- Optics Primer, Part 3: Co-Packaged Optics (CPO)
From EML lasers and DSPs to silicon photonics and external CW lasers. How CPO works and the impact on the optical supply chain.
- Optics Primer, Part 2: LRO & LPO
LRO and LPO move signal processing out of the transceiver and into the switch, saving power at the cost of tighter system coupling. An incremental step toward co-packaged optics.
- LiDAR, Explained: How It Works and Why It Matters
Autonomy, 905nm vs 1550nm, ToF vs FMCW, market dynamics, and more.
2025
- Optics Primer, Part 1: Traditional Pluggable Optics
Optical interconnects, pluggable transceivers, optical and electrical engines, DSP overhead, power and latency trade-offs
- SerDes Matters!
What Is SerDes, Why It Matters, Which Companies Have SerDes Chops
- Chiplets and the Future of System Design
Benefits and trade-offs, and the need for standards and an open ecosystem
- The Long Arc of Semiconductor Scaling
The journey from vacuum tubes to transistors, from small ICs to billion-transistor SoCs. Setting the stage for chiplets.
- Robots That See, Part 1: The Evolution of Computer Vision
Computer Vision, ConvNets, MNIST, ImageNet, AlexNet, Yann LeCunn, Fei-Fei Li
2024
- Semiconductor Metrology
Measure What Matters!
- Tokens Per Second Per Watt: A Useful Metric for Edge AI
A systems-level perspective on performance and energy efficiency
- High-Bandwidth Memory
What is HBM and why is it important for Generative AI?
- Memory: From Caches to Commodity Cycles
Memory Hierarchy, Memory Wall, HBM, Commodity Cycles, Historical Perspective, Beyond Silicon
- FPGAs - Speed of Hardware, Flexibility of Software
Deep-dive on FPGAs - Basics, Economics, Trade-offs, AI & LLMs
- ASICs: Custom Chips to Improve Performance
The benefits of ASICs. How custom silicon can make a company's "beer taste better". AI accelerators. The rationale behind Broadcom and Marvell's ASIC design service.
- GPUs: General-Purpose Parallel Computing
A Gentle Introduction to GPUs and Insights into Leading GPU Companies
2023
- CPUs: Hardware, Yet Infinitely Adaptable Through Software
A Beginner's Guide to CPUs and the Top Companies Shaping the Future
