bravura

'IRL paperclip maximizer: "Well first we should rob a bank..."'

That's too specific. Agentic AI learns subgoals that are generally valuable.

"Well let me learn to overcomb every jungle gym and if I cannot then to dissassemble the jungle gym and if that is not allowed to learn general techniques for avoiding cheating detection."

And this is where your analogy breaks down, because there are few natural analogues for the sorts of systems we are developing.

"Maybe we need very clear, standardized signals for entering a surveillance zone." [proceeds to give a libertarian pro-2A perspective and then the "what would a cab driver do perspective]

I have a suggestion. Perhaps if you enter an autonomous vehicle, it has a display with your name and will not drive until you buckle your seatbelt and click the button acknowledging?

You could also distill the fabulous Anticipatory Music Transformer from Stanford. https://crfm.stanford.edu/2023/06/16/anticipatory-music-tran...

Also consider checking their decisions about representation, etc

The PineTime watch unfortunately does not have any microphone, so there is no possibility of voice control.

"boring" means familiar. It should have more known unknowns than unknown unknowns.

What if the memory were git repo backed, and the FTS5 were a speed-specific optimization?

Then the memories could easily be human-reviewed. The repo would be the canonical source, and the FTS5 would be one specific materialization.

You sound like the sales rep at the Apple store. I was perplexed.

Meanwhile, I had to wonder how many people take showers that are over half an hour every day.

You just listed a bunch of 0 to 1 companies, not 1 to 10 companies.

They weren’t quantitatively better than previous companies. They were qualitatively better.

The important meta-conclusion that the article sadly misses is that the government will soon be using AI to regulate AI.

I still have my remarkable 1 and love it. RM 2 didn’t have a backlight so there was no upgrade reason.

I actually like that the RM 1 is plastic, it feels durable and light.

Not all liaisons are French, but all the dangerous ones are.

I’ve been curious for a while the cheapest way to create a small run of azulejo tiles with printed design and very strange (puzzle piece like) shapes

Close contact between exceptional individuals is one of the main values of having a scene.

"I like to write like a human, steering clear of jargon and B-school speak."

https://om.co/about/

https://om.co/2020/07/30/write-like-a-human/

Who taught you this? And why do you think this way?

A computer?

When you buy something in Germany and the EU, you typically can return it.

That does not apply to custom buildouts, like this camear.

It's a handful of useful features that together feel qualitatively different, like you're talking to a real person.

Believe me, I came here expecting to dunk on AI-haters, not talk tomato talk.

Yes. I didn't know or understand why I felt meh about run-of-the-mill raw tomatoes this until the Kenji article.

"Using mealy, off-season tomatoes is the primary unforgivable sin, but when it comes to BLT crimes, that's just the tip of the iceberg lettuce.

...

Off-season tomatoes are grown in warmer climates, picked when underripe, then treated with ethylene gas (a gas that is naturally produced by plants to trigger ripening in fruits) to produce their red color before they hit supermarket shelves. The result is tomatoes that are as bland as they are ruby-red.

Truly great tomatoes must be fully ripened on the vine, where they'll continue to develop flavor and sweetness. Look for plump tomatoes, with the heft and give of a water balloon. If you have a choice, look for substantial and meaty heirloom varieties with balanced sweetness and acidity, like Cherokee Purple or Brandywine.

Avoid tomatoes that feel light for their size, which means they have more air pockets inside and are typically better for saucing or salads."

Hot take: We mostly eat garbage tomatoes.

"*A BLT is a tomato sandwich, seasoned with bacon.*

It wasn't until I tasted my first great tomato, at the vine-ripe old age of 22, that I finally understood the true nature of the BLT (and, by extension, why I'd never enjoyed tomatoes on my sandwiches or in my salads). Here we go: A BLT is not a well-dressed bacon sandwich. A BLT is a tomato sandwich, seasoned with bacon. From this basic premise, all else follows."

https://www.seriouseats.com/ultimate-blt-sandwich-bacon-lett...

Bog-standard LLM mapping is terrible and I recently added Google Maps to my personal agent to remediate this.

I'd love to try Voygr for fun. Is there a skill defined that I could just swap in Voygr

You and GGP both wrong in ironic ways.

GGP says don't care about X because it's a social phenomenon, but frequently this position is a form of social identification.

You say: X might deeper than social, implying that social phenomena are not important. Thus agreeing with GP.

[edit: my position is pragmatic: If there's a broad or important phenomenon, your position on it should be individualized to the value of the phenomenon itself, not based upon some theory-of-origin category assignment.]

For me, trying to fine-tune a model to write "best day" prose I would accept over 80% of the time.

You are correct if we are talking about knowledge.

However it is bad at hyper-idiosyncratic, gritty style transfer.

I first noticed the issue when asking claude code to draft email responses. The choice of register was off. ("Register in writing refers to the level of formality and tone chosen to suit a specific audience, purpose, and context.")

I decided to talk all my HN comments and rewrite them in various bad LLM prose, and see if I could use DSPy to optimize a prompt using in-context-learning (ICL, I give it 10 examples of my HN comments) and the results were abysmal. RHLF fine-tuned frontier LLMs have a deep seated aversion to the target stylistic distribution of my comments.

I tried fine-tuning qwen3, llama, and gemma models. Instruct models are already so tuned that they could not be tuned. This is using several hunded comments as gold targets and 5 different LLM degradations per gold as the input.

How opinionated is this? Could it be a CLI?

Can it work with beads for tracking the state of everything?

i.e. do I need to buy in to an entire worldview to adopt this, or can I use it piecemeal?

My work partner set it up on telegram for himself and his wife and she uses it constantly. He was very surprised.

There are a few qualitative product experiences that make claw agents unique.

One is that it relentlessly strives thoroughly to complete tasks without asking you to micromanage it.

The second is that it has personality.

The third is that it's artfully constructed so that it feels like it has infinite context.

The above may sound purely circumstantial and frivolous. But together it's the first agent that many people who usually avoid AI simply LOVE.

Beeper API

"The point of MCP is bundling those exact things into a standardized plugin that’s easy for people to share with others." Like... a CLI/API?

"MCP is useful because I can add one in a single click for an external service" Like... a CLI/API? [edit: sorry, not click, single 'uv' or 'brew' command]

"So yeah, the agent could write some those queries manually" Or, you could have a high-level CLI/API instead of a raw one?

"I don’t get why people get worked up over MCP" Because we tried them and got burned?

"to help us get more context into agents in a more standard way than everyone writing a million different markdown files and helper scripts." Agreed it's slightly annoying to add 'make sure to use this CLI/API for this purpose' in AGENTS.md but really not much. It's not a million markdown files tho. I think you're missing some existing pattern here.

Again, I fail to see how most MCPs are not lazy tools that could be well-scoped discoverable safe-to-use CLI/APIs.