Replying to @zekramu
Much of it is trivial as a standalone eng concept.
However .. multi model, async agentic development, makes it less trivial.. especially when considering the agents insistence on hash identical validation.
Worktrees, limited compute, and storage, poor session planning, these all increase the complexity.
Then , with e2e integration testing, secrets management, VM setup, evidence capture, accessibility design principles, and other factors enter... then infra becomes the most challenging aspect of an agentic software factory.
Replying to @zekramu
The aspect of factory building that interests me is the validation, CI , E2E, VM , deployment process... Which has become the most difficult part of agentic development,
Replying to @kkoconnell
add it to memory.md, agents.md, claude.md, and create a hook that prevents a future pressing of button, maybe. Also, a skill may be useful here
Clint J. retweeted
There's not enough buzz around Claude Mods.
Read this Claude Mods article if you want to:
- Understand Claude Mods basics
- Install and use the most useful mods today
- Ponder where the agent-human interface will go
For @latentspacepod.
Replying to @izzyz
"From the remote depths of the corridor, the mirror spied on us. We discovered... that mirrors have something monstrous about them"
(Tlön, Uqbar, Orbis Tertius)
On Exactitude in Science Jorge Luis Borges, Collected Fictions, translated by Andrew Hurley. …
"In that Empire, the Art of Cartography attained such Perfection that the map of a single Province occupied the entirety of a City, and the map of the Empire, the entirety of a Province.
In time, those Unconscionable Maps no longer satisfied, and the Cartographers Guilds struck a Map of the Empire whose size was that of the Empire, and which coincided point for point with it.
The following Generations, who were not so fond of the Study of Cartography as their Forebears had been, saw that that vast Map was Useless, and not without some Pitilessness was it, that they delivered it up to the Inclemencies of Sun and Winters.
In the Deserts of the West, still today, there are Tattered Ruins of that Map, inhabited by Animals and Beggars; in all the Land there is no other Relic of the Disciplines of Geography. "
- Suarez Miranda,Viajes devarones prudentes, Libro IV,Cap. XLV, Lerida, 1658
Replying to @divinesturgeon @Windows
Probably they don't know any better.
Maybe they are too lazy to make meaningful change that will not show up immediately in their bonus.
Maybe, they are dumbs.
"In the elder days of Art,
Builders wrought with greatest care
Each minute and unseen part;
For the Gods see everywhere."
- Longfellow
Replying to @pravetz386
Thanks, I have been thinking about a fingerprint scanner that always outputs the same alphanumeric, but I'm not sure if it could be accurate enough with different quality scans on the same fingerprint.
Replying to @zekramu
Maybe there is a real weakness in agentic frontend design.. Maybe it is similar to the weakness in agentic creative writing.
Technically its right, but emotionally, aesthetically, contextually it lacks Soul.
I’m pretty disgusted with the product landscape. we have machines spitting out code and yet 99% of the interfaces I come across are basically stock anthropic ui shit. please spend some time making your product sexy, people like sexy things. the looksmaxxers were right about this.
Terrain Disclosure is when a map implies something before you’ve asked it anything.
Search answers the questions you already know to ask.
You type, it retrieves, and whatever you didn’t think of stays invisible.
A semantic terrain works the other way around.
Lay out a whole corpus by meaning, and the shape starts implying things on its own.
Two topics you assumed were separate turn out to share a ridge. Or there’s a valley where you’d have bet on a mountain.
You didn’t query for any of that. The terrain disclosed it, because it was there, and the previously unasked questions come next







