Assuming the author had to input good points for the LLM to expand it into a full length post, just give me the points. Give both formats if one must. Translate, spellcheck, fix mistakes with an LLM. Don’t inflate 3 bullet points to a 10 minute read.
I can’t distinguish between well meaning slop and blogspam slop. I might be able to if I’m well informed on the topic and can point out inaccuracies. Otherwise I just don’t know what is accurate and what is filler or outright wrong. LLM-isms are very distracting and exhausting to sort through. It just doesn’t make for a good reading experience.
Yes, this. I mean, if you're writing a novel meant to be read, maybe not. But if you're just trying to communicate 3 ideas, write them out in a way I can quickly tell if it's worth reading more.
There are few things (relatively speaking, I mean online activities, etc) more annoying than seeing a click-baity headline, and then having to read 17 paragraphs to actually get to the point that I realize I already knew and understood it, and it's just slop padding.
I have had a ton of success in exposing AST-based tools to agents when working in large, old codebases.
Without them, not only do agents get simple things like function call counts wrong, they tend to return different results. I use this as an example when showing people how the tooling works.
Grep is fine for simple use cases. A step up from that is ast-grep and I need to explore this tool more. But I had the most success building a small pipeline that reads the old code base, parses it file by file using tree sitter, and then loads it into a SQLite database for querying. For example, I have it capture construct definitions and usages and represent those as directed edges and nodes in a single table depending on the node type. The agent is instructed on how to query it and perform interesting queries like build call graphs, or determine dependencies between domains (modularity is not great in this codebase) which is helpful for us to extract around capability lines.
I also calculate fitness statistics, and have some code to capture specific details and knowledge about this very old framework that short circuits agent work in the future. We have some “interesting” magical libraries and functions that block static analyzers from going beyond the call site. This is mitigated, and means agents don’t have to “guess”.
Making all of this available to the different team members at my work has been pretty helpful. It’s faster (fewer tool calls), cheaper (fewer tokens), and accurate.
I can’t distinguish between well meaning slop and blogspam slop. I might be able to if I’m well informed on the topic and can point out inaccuracies. Otherwise I just don’t know what is accurate and what is filler or outright wrong. LLM-isms are very distracting and exhausting to sort through. It just doesn’t make for a good reading experience.