Adapter Boundaries That Make Change Cheaper
How to design adapter boundaries that contain vendor semantics without hiding business-critical differences or creating a private framework.
Read article →The field journal
Software Development
Practical architecture decisions, tradeoffs, and operating lessons from software that has to survive business change.
Browse all topics →Find the right article
Showing 5 of 34 articles
Latest writing
More practical architecture writing
One Model, Many Conversations: LLM Serving, Memory, and Reasoning
How serving engines share one model across users, isolate KV caches, batch token generation, preserve context, and spend more inference compute on reasoning.
From GGUF to GPU: How llama.cpp Actually Runs a Model
Trace llama_decode through GGML graphs, CUDA kernels, GPU memory, and next-token generation—and see why expensive LLM inference can still be fast.
An LLM Is Not the Program: What Is Actually Inside the Model File
Follow an LLM request past the API and into GGUF tensors, learned parameters, attention, and the transformer layer that turns model data into logits.
Build Around Business Capabilities, Not Vendor Features
Why durable business capabilities should define platform boundaries, while commerce, payment, and ERP vendors remain replaceable implementation details.