ai-engineering-systems-canon

Volume 4 — Runtime Architecture and Inference Mechanics

How AI systems actually execute under physical, computational, and operational constraints.

Reports

• AI-ENG-J — Throughput Mechanics: Memory Pressure, KV Cache & Hardware Realities
• AI-ENG-K — Weight Dynamics: Quantization, Compression & Decoding Strategy
• AI-ENG-L — Model Serving Architecture: Routing, Load Balancing & Deployment Topologies

← Back to Canon Map

This site is open source. Improve this page.