Dex Horthy argues context engineering, not model choice, determines AI-assisted code quality
The Pragmatic Engineer·medium signal
The Pragmatic Engineer interviewed Dex Horthy on why context engineering is the lever for building effective AI-assisted software without sacrificing code quality. The through-line is that teams blaming model capability for bad AI output are usually looking at a context problem — what gets loaded, in what order, and what gets evicted — which is directly actionable for anyone running coding agents daily.