I wish it were. I don’t like politics or ideologies.

Surprised this is not on substack. Saw another good writeup on video as a gist yesterday

this reminds me of the MIT billion prices project

I happened to go to brunch at a fancy restaurant on New Years Day. They were still updating their wine inventory with a spreadsheet.

Spreadsheets can work but get clumsy really fast. Bottle cost can vary as costs change (case discounts, price changes.. and holding inventory of an item across multiple purchases), audit logging (spelunking into the undo history), multiple staff interacting with the system.. All solved problems, just not at the price point or UX we need at this scale.

do you have a write up on your setup?

I sort of see it on the flip side. If you read through the MCP spec, there is the potential for the human input. If should be the AI doing all the grunt work it is capable of with the human putting in judgement when needed to complete some task.

I like to use AI as a sort of brainstorming tool. For example, if I have an idea for a software system, I will describe the general idea and ask for three different possible designs. I will tell it not to generate code but to provide three ideas with pros, cons, and tradeoffs. What it provides will often trigger some new ideas that take me off in a different direction. I often mix and match aspects of the ideas it provides with my own. I keep iterating on this till I think I have a design I like.

This is a great write up. Thank you for sharing.

I am hoping things revert to the mean and by extension, the consumer chip prices revert.

with investors and hype, things regress to the extreme

I think those that are most successful at creating maintainable code with AI are those that spend more time upfront limiting the nondeterminism aspect using design and context.