Advanced Packaging & The Chiplet Era: CoWoS, 3D Stacking, and Interposers
How breaking the reticle ceiling reshaped computing: TSMC CoWoS-L, silicon interposers, High Bandwidth Memory (HBM), and 10 TB/s chiplet interconnects.
Read Post →Featured Projects
View All Projects →Latest Blog Posts
Read All Posts →Streaming Multiprocessors (SM) & Compute Units: Schedulers, Registers, and Warps
Demystifying GPU execution mechanics: 32-thread lockstep warps, zero-overhead hardware schedulers, the branch divergence penalty, and Little's Law latency hiding.
Read Post →Moonshot AI Kimi K3.1 Pro: 2.8T Mixture-of-Experts, Dual-Attention, and 1M Multimodal Reasoning
Deconstructing Moonshot AI's Kimi K3.1 Pro: 2.8T MoE with 104B active params, 69 KDA + 24 Gated DSA attention layers, 1M context, MoonViT-V2, and quickstart guide.
Read Post →