jmartrican

> You can set up custom instructions to be less verbose and less jargon dense. You seem to be confusing the default RLHF personality with the inherent nature of AI. It will write however you ask it to write.

I agree. Its a skills issue. Get good at using your new LLM tool.

I agree, they shouldn't just proxy the Claude output directly to the SWEs. THey can use Claude all they want, but the output of their effort (which goes to the SWEs) should be understood (maybe even written by) the Ops engineer. The Ops eng can use Claude to dumb it down for them to understand and can even be taught by Claude on the issue further... so Claude to the rescue still.

The mental model I like to use is something along the lines of the LLM is a firehose and my brain is a mere straw and that we have to eventually get all the important parts to trickle through my straw sized brain. Or another way, the LLM is all the dough shoved into a pasta extruder and all the dough needs to eventually extrude out of my brain into prod-pasta.

Very cool. How have you liked the results of doing this? Any pitfalls so far with this approach?

Prediction of a future transcript at a press conference after some major AI caused disaster: "We had no idea that the model would be capable of..... ".

The problem is picking between models. I do not want to spend my time switching models and trying to decipher which model should be used for what. Maybe that's just a me problem that I need to figure out.

What if we remove the stigma of authors using AI. We say, have at it, use all the AI you want. What then? The market is still finite. The authors are still competing with each other. They will still have to be good at creating compelling books that people want to read. Will the whole thing just blow over and AI would be "just another tool" that authors use to get their work out?

I think the name of the game is going to be "how do we maximize output while keeping token counts low". For example, finding ways to replace AI workflows, even if minor ones, with scripts/code.

Anthropic told me that I have till August 19th to be get my act together becuase they are going to reduce my token count by 50%. lol. I have been abusing my Pro Max allowance and need to start being less wasteful.

Articles like this, can help us come up with ideas on how to do it.

The author said that Claude was not good at refactoring. But the author was using Sonnet 5. I supect Opus 5 would have done a lot better at refactoring. It would have come up with a multstep approach, i suspect.

I like Bill's initial response to Steve... "No, there's no way to do that. In fact it would be really hard to do, and I don't think we really need it". Classic developer response when their code is called out. lol

Regex came up so infrequent that I found myself referring to documentation whenever I needed to use it. But I always wondered, what are the jobs or roles that use it so often that they have mastered it.

I think that the onus is on us to get better at using agents and AI to solve the pain points and speed things up while keeping quality high and our mental tools sharpened. I do nto think turning back is an option, but managing the pain points and leveling up is.

> So what’s the ideal “middle ground” in this situation?

I use agents to code. But I remember the early days of just AI smart complete in the IDE, where as the programmer I had to be more involved with designing and implementating the solution. This kept me engaged with the implementation as it was being built out. Now with agents, I find myself trying to catch up with what the agent did and spend more time code reviewing. Maybe you end up in the same place in the end. But building the implementation, vs code reviewing, feels more rewarding and I think helps keep your mental tool sharpened.

Oh yeah, i should have guessed. Thanks for pointing it out.

IMHO they should cosnider re-releasing the 3310. It might sell similar to vinyl. Maybe update the games section to have some more games.

The concern would be the ever increasing cost of GPUs. I imagine to help mitigate the risk, having a setup where more GPU's can be added on, and the old ones can still be used (as new ones get added on) would make the endeavor more palatable. Vs. having to swap out old GPUs for new ones, because the money spent on the old ones is lost (or at least some of it is lost if you can find a way to resell them).

Sounds expensive

Ford Authority also covered this story: https://fordauthority.com/2026/06/ford-unified-operations-hi...

Do you have any references for this comment: '"water use" thing is a myth anyway'. I would like to show my kid who keeps complaining to me about AI water usage.

I worked for a small successful company in the pharmaceutical industry. One of their founders died in a crash in his small personal aircraft. He was a really nice guy and very charismatic. I was not working there when the accident happened, but I was sad to hear about it.

I agree with OP's sentiment.

Maybe that is a skills issue.

Oh wow I didnt even know about that. Yeah it auto switches. I'll change the config.

I have the $100 plan and now quickly get downgraded to Sonnet. But so far have not hit any other limits. I use it more on the weekends over several hours, so lets see what this weekend has in store.

I suspected that something like this might happen, where the demand will outstrip the supply and squeeze small players out. I still think demand is in its infancy and that many of us will be forced to pay a lot more. Unless of course there are breakthroughs. At work I recently switched to non-reasoning models because I find I get more work done and the quality is good enough. The queue to use Sonnet 3.7 and 4.0 is too long. Maybe the tools will improve reduce token count, e.g. a token reducing step (and maybe this already exists).

reminds me of Snow Crash, where the US federal government is reduced to a software entity.

Is the book any good? Like is the story/poems in it worth reading? Just asking cause I might want to read it if its good.

I worked in a backend team that introduced JWT, before I got there. The problem we had with JWT was that the data was stale. Even if it wasn't stale, it needed to be treated as stale because every service wanted the up to date data, even within 1 sec that data is old. The user could have changed something in their account from the time that the JWT was issued. I removed JWT and went back to the old UUIDs as tokens.

This risk with using the simple tech to start out is that you are one feature request away from being under engineered and will have to rebuild the whole thing.

The funny thing about the 3D chips is that if extrapolated far out, we end up with cubes. Definitely sci-fi territory. And maybe questions from kids like "but if they are shaped like cubes, why are they called chips?".

> Currently TSMC has the only leading edge chip fabrications plants (fabs) on the planet and they're all located in Taiwan.

What about Samsung? I thought they also made leading edge.

To think how many things occurred during 1250 to 1860. All that progress was made during the little ice age.