Project Vend: Can Claude run a small shop? (And why does that matter?)

This sounds like they have an LLM running with a context window that just gets longer and longer and contains all the past interactions of the store.

The normal way you'd build something like this is to have a way to store the state and have an LLM in the loop that makes a decision on what to do next based on the state. (With a fresh call to an LLM each time and no accumulating context)

If I understand correctly this is an experiment to see what happens in the long context approach, which is interesting but not super practical as it's knows that LLMs will have a harder time at this. Point being, I wouldn't extrapolate this to how a commercial system built properly to do something similar would perform.

In my experience long context approach flatly doesn't work, so I don't think this is it. The post does mention "tools for keeping notes and preserving important information to be checked later".

From the article:

It had the following tools and abilities: * Tools for keeping notes and preserving important information to be checked later—for example, the current balances and projected cash flow of the shop (this was necessary because the full history of the running of the shop would overwhelm the “context window” that determines what information an LLM can process at any given time);

Yeah it's not clear

> The shopkeeping AI agent...was an instance of Claude Sonnet 3.7, running for a long period of time.

This is what made me wonder. What does running for a long period of time mean? Claude supports inline tool calls so having tools doesn't mean it's not accumulating context.