They use bots to promote their content to the front page and downvote any comments criticizing them. I observe this pattern consistently on this website.
You're right. This was just a performance issue, but what if next time it is a corruption bug or a security vulnerability or really anything that can cause real consequences if happened in production? I don't think that LLM systems are inherently bound to have this flaw, but I think that we are pretty far from harnesses and algorithms becoming advanced enough so that the LLM system can kind of continuously evaluate its output and ensure it is good in all aspects.
It has good test coverage, mostly unit tests but also a number of end-to-end tests. I also made the LLM build a benchmark, which you can find at the bottom of the readme. It is obviously slow, but I thought that it is good enough to work. When I tried to write a 1 GiB file, I found that it broke down, and after writing half the file, the speed went to under one megabyte per second. Implementation is 10k+ LoC, and I have no idea what is going on there.
> as does python, which had a reputation for decades as THE language to use to teach new folks to code
I am very perplexed by this. I am going through Neetcode's DSA course where he explains what RAM and arrays are, but then he goes on to say something like "but since we are going to use Python, none of this applies." Personally, I learned the most about how software really works from reading The Rust Programming Language. It not only teaches you how to program in Rust, but also how memory works, what a string really is, etc.
I think in this context it is actually important to share exact wording that causes the AI to perform well. My favorite is "Do not use your own knowledge."
Why did you opt for semantic search, and not plain old full text search? I built an "AI Agent for a Commerce Website" as a take-home exercise yesterday, and I chose to simply give the model a tool that does a full text search over products, powered by MiniSearch, and I think it works reasonably well. I believe this is also what Claude Code does.