AI Engineering

2 expert articles in the AI Engineering category — practical knowledge and tips for your web project.

Benchmarking qwen3-coder locally building an Astro 6 site on MacBook Pro M4 Max
Ollama Local LLM qwen3-coder Astro 6 Claude Code AI Engineering Benchmark Apple Silicon

Benchmarking local qwen3-coder on an Astro 6 build on M4 Max — 4 model failures, not Mac failures

Honest benchmark: 64 GB unified memory, 40 GPU cores, qwen3-coder:30b on Ollama, Claude Code as the harness. The model confused Astro with Svelte, ignored an explicit STOP, and stalled 30 min on a single turn. Technical diagnosis + the real fixes.

12 min read
Read more
Diagram showing a constitution.yaml feeding into a Python executor with a halt-on-reject gate
LLM Agents Python Claude Code Anthropic Policy Open Source

LLM Agents Do Not Respect Rules. This Pattern Makes Them.

Plan Mode reviews proposals, not execution. Reference implementation of the executor-side halt-on-reject pattern with Kustomize-style declarative rule merging.

9 min read
Read more