My tests in the past month, basically $0 saved over my findings: - Two one-way flights - New OLED 55"- 65" TV - Mini PC for CI/CD runner, with very specific specs - Expensive, specialized medical device - Variety of car parts

I think one problem is that I simply already know the places to check for this stuff. The more esoteric places (ebay/fb marketplace/etc) agents seem to just be awful at parsing what is real, what is a scam, what is a good deal, etc, and that is where the big deals are for stuff.

Complications with healing are pretty brutal. Necrosis, amputation, infection are all common without the built-in cascade we have.

If the tool won’t be upfront about those things, they are not worth looking into any further. It’s used car salesman strategy.

Jobs, internships, networking, and software engineering courses are critical. They let you speak like someone with much more experience. You don’t want to come into a questionable market with only raw CS skills. You will learn the CS, that is a given by your degree, so hone the critical people skills and how to work in a team. In a downturn no one wants to hire an unknown, so you need to be able to show that you aren’t green.

Internships with companies that are known to hire their interns are also a very good idea, even if it’s a “boring” company. These can be competitive, you gotta put that effort in.

Another good route in your later college years is also underpaid work assisting grad students or professors doing grant work. It shows you can apply what you learned, no one really wants to do it, it builds professional relationships, and it’s real job experience.

Extra prompt text isn’t to make the model smarter, it’s to pull attention towards what you want. “Do code review” is far different from “Make sure changes align with existing architecture” or “follow these enterprise standards XYZ”. The model just acts as an average of its training data, which may or may not align with your goals.

1. We already know the headline cherry picked savings will not be true in real world agentic coding. Why not approach this more scientifically? The salesman/hype thing is played out in this space imo. 2. If these tools really save some double digit percentage of tokens at no downside, why has no provider rolled this out by default yet to undercut all their competition?

Instead, it is my experiences with LLMs in a domain that I know very well that makes me skeptical of their performance across the board. I find issues in code review multiple times a day with their output, and they are explicitly and extensively trained on this use-case, unlike with the MRI data. Sometimes I veer into other domains I have decent knowledge about (construction, carpentry, landscaping) and LLMs disappoint me there as well.

I suppose Gell-Mann amnesia is a universal human quirk and not restricted to just the news.

Wal-Mart is the opposite. I go in to buy groceries and am always astonished to have my trip be like for $27 or something. I usually shop from the outside of the grocery (raw produce) and they are just dirt cheap.

HEB operates a series of upscale Central Market stores that are even more lopsided in price and even more interesting in selection.