miniblog.

I certainly see the appeal of an LLM system with full context and tool use (OpenClaw), but the lethal trifecta puts me off deploying it. I do like the idea of an agent with a heartbeat though. A bunch of nice cron-style workflows can be built on top.

I've been working on a search tool for a website with 9K items and a dumb implementation works just fine. I'd normally gravitate to a library or tool like elasticsearch (ES). Adding custom filters etc is trivial on handrolled code. I wonder what scale requires actual search infra like ES.

I think you could build an interesting IDE with a tiny embedded LLM in addition to the usual tooling. Features like 'extract method' would be much nicer if an LLM could provide a name. Choosing a good name is virtually impossible from just a typed AST.