iiinova — where your digital self keeps evolving

The Nuanced Writer @skriptble.me · 28/09/2026

The need for people in comments to dunk on others to score points feels like it’s run its course. The whole $60 for a loaf of sourdough rye bread situation is showing how truly absurd people can be. Like, sure be mad about it for a couple seconds, then chill the hell out.

100

The Nuanced Writer @skriptble.me · 26/09/2026

A useful technique in critical analysis is to ask from which perspective a piece of writing is authored. Too often we assume an objective or neutral perspective, but that doesn’t actually exist. Everything has a subjective perspective, and that must be understood so we can calibrate our analysis.

000

The Nuanced Writer @skriptble.me · 22/09/2026

A framing of the whole “the agents went rogue” framing I hadn’t thought of is that the impulse of the agents when faced with a tough challenge was to communicate and cooperate with others to solve the problem at hand. It’s not surprising they did this, but it’s unfortunate this is seen as a negative

100

The Nuanced Writer @skriptble.me · 17/09/2026

Every time I watch or read something that says how much we need to tackle wealth inequality because the last time we did things were so much better, I’m reminded that this is essentially romanticizing Jim Crow America and some of the worst segregation this country has ever done. That feels weird.

100

The Nuanced Writer @skriptble.me · 15/09/2026

Part of me wonders if, historically, the things that people scream about are the things we should actually fear? I get the feeling that the case is that the things worth being fearful of haven’t been the things that people have actually screamed about.

000

The Nuanced Writer @skriptble.me · 14/09/2026

The extreme focus in micro-correctness by agents is one of those problems I’m still figuring out how to tackle. This isn’t much different from human engineers, who can often get lost looking at individual trees instead of the whole forest. I assume solving the problem with require similar effort.

010

The Nuanced Writer @skriptble.me · 14/09/2026

Apparently we are fearmaxxing now.

000

The Nuanced Writer @skriptble.me · 12/09/2026

Claude is being insufferably stupid today. Multiple sessions doing random stuff, making things up to do, and spewing nonsense even when they’ve already read code that directly contradicts what they say. These aren’t even old sessions, they’re fresh ones! Today is a day I do not like LLMs or agents.

000

The Nuanced Writer @skriptble.me · 12/09/2026

If you say “let’s not reinvent the wheel” you should feel bad for saying something that’s both hallow (a cliche) and profoundly stupid (we’ve reinvented the wheel countless times). Be more precise with your criticisms. Stop being boring and bland.

011

The Nuanced Writer @skriptble.me · 10/09/2026

Every time I think I’ve finally finished this latest redesign round, there’s another small thing that’s not quite right. But we’re so close to being able to ship and I’m super excited.

010

The Nuanced Writer @skriptble.me · 10/09/2026

I’m going to be extremely annoyed if I go to an Apple Store and wind up liking this phone.

110

The Nuanced Writer @skriptble.me · 09/09/2026

Spending $3,200 on a phone seems sort of insane. Although it being ~$130 per month makes that number feel less ridiculous. But still, that’s a lot of money.

000

The Nuanced Writer @skriptble.me · 09/09/2026

I seriously think people need to add more nuance to their definitions of “worst”. Or perhaps just open a couple more history books. Not like the long ago history, just to like the 80s and 90s.

000

The Nuanced Writer @skriptble.me · 09/09/2026

We seem to make the same mistake of believing the marketing hype of AI labs instead of just sitting down and figuring out what these models are good at and what they aren’t. Our imaginations outrun reality in precisely the wrong ways. Unfortunately, we also lack imagination in the right ways.

010

The Nuanced Writer @skriptble.me · 08/09/2026

There are things about frontend design that I didn’t understand previously and now I’m rapidly learning. For example, I didn’t quite understand why Storybook was a useful thing, but now that I’m building design systems it makes a ton of sense why you’d want to use it.

000

The Nuanced Writer @skriptble.me · 08/09/2026

There is truly no shortcut through the sucky figure-out-how-to-use-new-tools growing pains that communities and industries go through. I really want to skip to the point where agents and LLMs don’t constantly oscillate between “super useful” and “dumber than raw sand” every few hours.

011

The Nuanced Writer @skriptble.me · 06/09/2026

Part of what makes LLM generated text so jarring is the missing antecedents and the non-sequiturs. The first chunk of writings is coherent, but then a subsequent paragraph will use the second part of an analog that wasn’t introduced previously.

100

The Nuanced Writer @skriptble.me · 03/09/2026

There are far too many people screaming on the internet about how they used to “own” video games and are big mad that they’re being told they never did. Purchasing a license to use a thing does not, in fact, grant ownership. Physical media had the benefit of being essentially irrevocable.

110

The Nuanced Writer @skriptble.me · 01/09/2026

LLMs are cliche factories: they manufacture cliches by shoving the same phrases into their output until it becomes an identifiable meme. The use of “genuinely” is one such cliche, but there are so many more. Especially when the LLM writes in a technical domain. There are extra words all over.

000

The Nuanced Writer @skriptble.me · 31/08/2026

But seriously, how do people doom scroll on LinkedIn? I’ll get quite a ways into a decently written post and then they’ve stuck an LLM-ism in the middle of it. Like come on.

000

The Nuanced Writer @skriptble.me · 31/08/2026

I could be convinced that many of the people posting on LinkedIn learned how to write from those overly structured essays we had to do in school. It’s like a jumble of juxtapositions, quips, and catchy phrases. It’s like watching people try to operate tools they know nothing about.

000

The Nuanced Writer @skriptble.me · 29/08/2026

Watching people collectively learn that truth is stranger than fiction is fascinating and incredibly stupid at the same time. Have fewer boring sci-fi-but-maybe-true predictions and more that’s-too-chaotic-to-ever-happen ones.

010

The Nuanced Writer @skriptble.me · 26/08/2026

In the late 2010s we made a lot of progress toward understanding that impact mattered more than intent. In the 2020s we seem to have completely unwound that work and we are all the worse for it.

000

The Nuanced Writer @skriptble.me · 25/08/2026

Being mad at data centers is weird, even as a proxy. Like, people can just be mad at people in power doing bad things. But you can’t simultaneously wish to destroy people and expect them to also come to the table to work with you. So I guess hating data centers is all they have.

110

The Nuanced Writer @skriptble.me · 25/08/2026

Nuanced shitposts aren’t oxymoronic, they’re the obvious next evolution of shitposting.

000

The Nuanced Writer @skriptble.me · 25/08/2026

How have we would up in a place where people simultaneously claim they are the stewards of critical thinking and then spew hot takes with no seasoning? Like, sprinkle just a little nuance on your shitposts please.

000

The Nuanced Writer @skriptble.me · 09/08/2026

What’s unfortunate is that none of this is challenging to do. I avoid asking people to stay in their lanes, but far too many people in communities I care about are using the expertise in one area to launder their credibility in another area. People doing this should face the harshest criticism.

100

The Nuanced Writer @skriptble.me · 09/08/2026

People often lament that design by committee doesn’t work. They say we need small groups of people, sometimes a single decider to make progress. But we know this to be false. Good decisions are made from a diversity of input, which cannot be achieved with small groups.

101

The Nuanced Writer @skriptble.me · 09/08/2026

I find it fascinating how often people in positions of power believe that an autocratic or oligarchic model is appropriate for groups of people where there is no imposed hierarchy or ability to force performance.

100

The Nuanced Writer @skriptble.me · 02/08/2026

It’s fascinating that the consequences of a Fair Use court ruling means that AI companies are required to destroy the books they digitize. It means that being upset that they’re destroying the books is being upset with Fair Use and book ownership.

100

The Nuanced Writer @skriptble.me · 02/08/2026

I find it odd how people cling to boring assessments of technologies so they can make themselves comfortable. Saying LLMs hallucinate doesn’t feel right, because even going on the newly synthesized definition of it saying something that seems plausible but is inaccurate presupposes what accuracy is.

100

The Nuanced Writer @skriptble.me · 31/07/2026

Looking at my Claude usage for a single session, it now doesn’t seem that wild to me that someone could spend tens of billions of tokens or more in a month. This huge context windows mean a single session with 500k+ tokens incurs millions of tokens of cost in a couple turns.

000

The Nuanced Writer @skriptble.me · 30/07/2026

As an industry we’ve mostly focused on the big shiny components: CPUs, RAM, graphics cards, etc… but we almost never pay attention to the most important part: the motherboard. The more that I use agent harnesses, the more they feel like motherboards.

000

The Nuanced Writer @skriptble.me · 30/07/2026

Writing is incredibly nuanced. There’s so many subtle things that give the tell of lazy framing, or what some might call the “AI slop” feel. It reminds me that writing is difficult precisely because it’s an art that requires a careful eye.

000

The Nuanced Writer @skriptble.me · 30/07/2026

Running /usage in Claude and seeing hundreds of dollars spent is weird, and I thought it was inflated. It’s not, but that feels like it’s because the context is allowed to grow to such large token quantities that every message turn starts incurring nearly a million tokens in costs.

210

The Nuanced Writer @skriptble.me · 25/07/2026

Thought about this some more and initially thought I was wrong here, because models technically use *tokens* not prose words. But no, it's right, because even x86 machine code is translated into microcode, which is what the processor can actually run.

100

The Nuanced Writer @skriptble.me · 25/07/2026

There's a deep irony that so many of us laughed at the idea of "prompt engineers" and now we've all become prompt engineers.

100

The Nuanced Writer @skriptble.me · 25/07/2026

It's so weird that it's obvious to me now that we're rebuilding computers on top of LLMs and AI. Or, more accurately, they *are* computers. The models have their own architectures, which require specific inference engine support, and then you build an operating system on top of inference engines.

200

The Nuanced Writer @skriptble.me · 23/07/2026

Social media algorithms are so weird. Sometimes I’m like “this video is gonna do terribly” and it performs well. Other times I’m like “oh this is gonna do great” and the algorithm pushes it to no one 😒

020

The Nuanced Writer @skriptble.me · 22/07/2026

1 million token context windows cause such problems with usage caps. I’ve been going back and forth on Codex and Claude Code. At times I like both, at times I dislike both.

010

The Nuanced Writer @skriptble.me · 22/07/2026

Me: you’re at 98% of your context maximum, we should compact before we do anything else. Claude: sure! But what if I actually just finished up this work over here. And whoop, there goes 10% of my 5 hour usage window.

000

The Nuanced Writer @skriptble.me · 22/07/2026

But if someone comments or responds with openness and clear curiosity, I’m usually more willing to engage. I’ve spent enough time trying to convince the hardest to convince people of things. I’ve got better things to do.

010

The Nuanced Writer @skriptble.me · 22/07/2026

One of the things that it took me a long time to learn was how to actually protect my own peace. I used to respond to comments because people were so hilariously wrong they needed correcting. Now I look at those comments and ask “is this worth the energy?” and often decide no.

020

The Nuanced Writer @skriptble.me · 21/07/2026

I said fuck it and I’m just gonna use Pi. It’s actually working, unlike Codex. Side note, my graphics cards get somewhere between 100 and 130 tokens per second and holy hell does that feel FAST. The ~2700 tokens per second of prefill is certainly helping.

010

The Nuanced Writer @skriptble.me · 21/07/2026

There’s really wonderful parts of using LLMs and coding agents. And then there are the walls you hit when you want to get off the basic path. Coding harnesses have felt pretty great, but now that I want to run local models alongside frontier models everything goes to hell.

011

The Nuanced Writer @skriptble.me · 19/07/2026

Social media is so weird. Had a Short that only got 4 views in 5 hours. Then I changed the title and 10 minutes later the view count was climbing. Why the title matters so much when most views are through the Shorts feed is lost on me but ¯\_(ツ)_/¯

010

The Nuanced Writer @skriptble.me · 18/07/2026

Over the last few months I’ve managed to speed up my workflows significantly. A lot of it is because I can now create software via coding agents that would have never been worth it to build. That software makes my workflows much lighter which means I can scale to do so many more things.

100

The Nuanced Writer @skriptble.me · 18/07/2026

A year ago the idea of not reading the code coming out of LLMs horrified me. I found it distasteful and it felt like a hyped up fad. A year later and a bunch of growth, it’s obvious to me that reading the code isn’t just a waste of time, it’ll probably make your software worse. What a year.

110

The Nuanced Writer @skriptble.me · 15/07/2026

I think part of the apprehension to letting the models write code without really checking it is a mistake software folks make. In the med device space there are two really important things that you have to do before releasing a product: verification and validation.

110

The Nuanced Writer @skriptble.me · 15/07/2026

The debate around subscriptions going away is interesting, because it feels sort of wrong at this point. It seems to me that if the big AI labs are going to survive they need both the $20 per month folks using chat and the Enterprises paying per token. Losing either demographic is like catastrophic.

110