There are large areas of software where the quality of code is measured by the mental model of its human maintainers. For these areas, LLMs are a net negative because their use degrades the mental model of the human maintainers.
There are large areas of software where the quality of code is measured by the mental model of its human maintainers. For these areas, LLMs are a net negative because their use degrades the mental model of the human maintainers.
Even just reading the code, their ability to ingest colossal amounts & answer questions in incredibly useful. Once you hook them up to observability, DAP (debug anything protocol), a repl, devtools... their ability to augment and accelerate practitioners is amazing.
There can be a lot of oral history that is lost, for sure, that helps illuminate why things are built the way they are, what the intent was. I've always favored a written culture, that has artifacts that help explain the past. Engineers move projects, move companies: trusting the mental model to live in people's heads has a lot of risk and downsides to it.