Daniel Grant
@djgrant.bsky.social
Building stuff. Blog at danielgrant.co.
Reposted by Daniel Grant
There's growing experimental evidence¹˒²˒³˒⁴ that LLMs are unusually good at talking people out of entrenched false beliefs and conspiracy theories.
¹ pubmed.ncbi.nlm.nih.gov/39264999/
² pubmed.ncbi.nlm.nih.gov/41179846/
³ sciety.org/articles/act...
⁴ connecticut.adl.org/resources/re...
¹ pubmed.ncbi.nlm.nih.gov/39264999/
² pubmed.ncbi.nlm.nih.gov/41179846/
³ sciety.org/articles/act...
⁴ connecticut.adl.org/resources/re...
October 3, 2026 at 7:00 PM
There's growing experimental evidence¹˒²˒³˒⁴ that LLMs are unusually good at talking people out of entrenched false beliefs and conspiracy theories.
¹ pubmed.ncbi.nlm.nih.gov/39264999/
² pubmed.ncbi.nlm.nih.gov/41179846/
³ sciety.org/articles/act...
⁴ connecticut.adl.org/resources/re...
¹ pubmed.ncbi.nlm.nih.gov/39264999/
² pubmed.ncbi.nlm.nih.gov/41179846/
³ sciety.org/articles/act...
⁴ connecticut.adl.org/resources/re...
Reposted by Daniel Grant
the claude dining experience
July 14, 2026 at 5:43 PM
the claude dining experience
One shift I've noticed coding with agents: a greater appreciation that I am channeling the sum of human knowledge. I think this was always true – we stand on the shoulders of our ancestors – but now it's explicit, I am more humble about my contributions.
July 22, 2026 at 10:07 AM
One shift I've noticed coding with agents: a greater appreciation that I am channeling the sum of human knowledge. I think this was always true – we stand on the shoulders of our ancestors – but now it's explicit, I am more humble about my contributions.
contemplating "principle of least software"
a key source of slop is creating more software than needed to achieve its aim e.g. building an electron app with auth when a cli tool would do.
we do this because it is so cheap to create software. there is a lack of back-pressure.
a key source of slop is creating more software than needed to achieve its aim e.g. building an electron app with auth when a cli tool would do.
we do this because it is so cheap to create software. there is a lack of back-pressure.
July 4, 2026 at 3:27 PM
contemplating "principle of least software"
a key source of slop is creating more software than needed to achieve its aim e.g. building an electron app with auth when a cli tool would do.
we do this because it is so cheap to create software. there is a lack of back-pressure.
a key source of slop is creating more software than needed to achieve its aim e.g. building an electron app with auth when a cli tool would do.
we do this because it is so cheap to create software. there is a lack of back-pressure.
i've been preserving agent threads for when the thing they've built gets broken - just return to the original thread and tell it to restore the feature.
it got me thinking: what if i had a directory of these threads that served as a kind of regression suite?
it got me thinking: what if i had a directory of these threads that served as a kind of regression suite?
July 3, 2026 at 9:28 PM
i've been preserving agent threads for when the thing they've built gets broken - just return to the original thread and tell it to restore the feature.
it got me thinking: what if i had a directory of these threads that served as a kind of regression suite?
it got me thinking: what if i had a directory of these threads that served as a kind of regression suite?
Reposted by Daniel Grant
RP2350 as the MCU co processor on the Flipper one 👀
Flipper One — we need your help
With Flipper One, we're reimagining what a Linux cyberdeck can be — it's a huge project. We're opening up the development process and asking the community for help.
blog.flipper.net
May 21, 2026 at 2:03 PM
RP2350 as the MCU co processor on the Flipper one 👀
Reposted by Daniel Grant
🎉 TinyBase v8.3 - the one with @solidjs.com!
Primitives and components to work with Solid - just like you can with React & Svelte!
Major props to Daniel Grant (@djgrant.bsky.social) for the hard work to get this going!
Primitives and components to work with Solid - just like you can with React & Svelte!
Major props to Daniel Grant (@djgrant.bsky.social) for the hard work to get this going!
May 9, 2026 at 1:56 AM
🎉 TinyBase v8.3 - the one with @solidjs.com!
Primitives and components to work with Solid - just like you can with React & Svelte!
Major props to Daniel Grant (@djgrant.bsky.social) for the hard work to get this going!
Primitives and components to work with Solid - just like you can with React & Svelte!
Major props to Daniel Grant (@djgrant.bsky.social) for the hard work to get this going!
Talking to a model that seems eerily human? BUG!
This is very similar to the skeuomorphism trend that accompanied the launch of the iPhone. And once it has served its purpose it needs to be burned with fire.
substack.com/home/post/p-...
This is very similar to the skeuomorphism trend that accompanied the launch of the iPhone. And once it has served its purpose it needs to be burned with fire.
substack.com/home/post/p-...
Anthropomorphism is the New Skeuomorphism
This phase needs to be brough to a swift end.
substack.com
May 6, 2026 at 8:41 AM
Talking to a model that seems eerily human? BUG!
This is very similar to the skeuomorphism trend that accompanied the launch of the iPhone. And once it has served its purpose it needs to be burned with fire.
substack.com/home/post/p-...
This is very similar to the skeuomorphism trend that accompanied the launch of the iPhone. And once it has served its purpose it needs to be burned with fire.
substack.com/home/post/p-...
conveniently, the market for people who think they have good taste is quite large
April 1, 2026 at 2:06 PM
conveniently, the market for people who think they have good taste is quite large
these two positions are irreconcilable:
1. tech debt doesn't matter
2. context is really important
tech debt *is* context! code is an ontological representation of the thing you are building. if it's not coherent, you are literally feeding your agent a million bad prompts
1. tech debt doesn't matter
2. context is really important
tech debt *is* context! code is an ontological representation of the thing you are building. if it's not coherent, you are literally feeding your agent a million bad prompts
March 19, 2026 at 12:58 PM
these two positions are irreconcilable:
1. tech debt doesn't matter
2. context is really important
tech debt *is* context! code is an ontological representation of the thing you are building. if it's not coherent, you are literally feeding your agent a million bad prompts
1. tech debt doesn't matter
2. context is really important
tech debt *is* context! code is an ontological representation of the thing you are building. if it's not coherent, you are literally feeding your agent a million bad prompts
those who *rely* on AI for guidance will have a painful collision with reality
March 12, 2026 at 3:35 PM
those who *rely* on AI for guidance will have a painful collision with reality
Reposted by Daniel Grant
Annnd the blog posts have been published! 🐝 Find them on my website in the little writing nook 💖
jclahoot.com?position=%7B...
jclahoot.com?position=%7B...
March 9, 2026 at 7:42 PM
Annnd the blog posts have been published! 🐝 Find them on my website in the little writing nook 💖
jclahoot.com?position=%7B...
jclahoot.com?position=%7B...
Reposted by Daniel Grant
I wrote about some things I'm struggling with in the technology industry, preceded by an appreciation for folk music.
I am in an abusive relationship with the technology industry
Some things I'm struggling with in the technology industry, an appreciation for folk music, and some other thoughts.
whitep4nth3r.com
March 6, 2026 at 3:37 PM
I wrote about some things I'm struggling with in the technology industry, preceded by an appreciation for folk music.
how quickly we regressed from not reviewing the code to not reviewing the strike target
March 3, 2026 at 1:29 PM
how quickly we regressed from not reviewing the code to not reviewing the strike target
Claude stuck fixing a bug.
No problem. I can deal with this.
Roll up sleeves. Open codebase.
Oh good god
No problem. I can deal with this.
Roll up sleeves. Open codebase.
Oh good god
February 27, 2026 at 3:04 PM
Claude stuck fixing a bug.
No problem. I can deal with this.
Roll up sleeves. Open codebase.
Oh good god
No problem. I can deal with this.
Roll up sleeves. Open codebase.
Oh good god
Like a siren, the closer you get to an LLM, the harder it becomes to peceive what it actually is.
February 26, 2026 at 9:38 PM
Like a siren, the closer you get to an LLM, the harder it becomes to peceive what it actually is.
Codex 5.3: "This epic improves review throughput with concrete fixes, not aesthetic refactors"
Me: "That is a bullshit statement"
Codex 5.3: "Agreed. I'II replace it with concrete, non-bullshit wording and tighten the whole epic to explicit problems"
Me: "That is a bullshit statement"
Codex 5.3: "Agreed. I'II replace it with concrete, non-bullshit wording and tighten the whole epic to explicit problems"
February 24, 2026 at 2:26 PM
Codex 5.3: "This epic improves review throughput with concrete fixes, not aesthetic refactors"
Me: "That is a bullshit statement"
Codex 5.3: "Agreed. I'II replace it with concrete, non-bullshit wording and tighten the whole epic to explicit problems"
Me: "That is a bullshit statement"
Codex 5.3: "Agreed. I'II replace it with concrete, non-bullshit wording and tighten the whole epic to explicit problems"
thinking about why openclaw blew up
1. people love talking to agents with a persistent personality
2. perhaps the first maintstream example of a self-improving software environment
the personal assistant part wasn't new
1. people love talking to agents with a persistent personality
2. perhaps the first maintstream example of a self-improving software environment
the personal assistant part wasn't new
February 19, 2026 at 4:32 PM
thinking about why openclaw blew up
1. people love talking to agents with a persistent personality
2. perhaps the first maintstream example of a self-improving software environment
the personal assistant part wasn't new
1. people love talking to agents with a persistent personality
2. perhaps the first maintstream example of a self-improving software environment
the personal assistant part wasn't new
if everyone has a personal agent, collective action becomes a much less tractable problem. people already have aligned goals; it's the coordination that's tricky. agents can probably help here.
that could have consequences.
that could have consequences.
February 19, 2026 at 4:31 PM
if everyone has a personal agent, collective action becomes a much less tractable problem. people already have aligned goals; it's the coordination that's tricky. agents can probably help here.
that could have consequences.
that could have consequences.
amazing how fast we transitioned from "everything should be rewritten in rust for memory safety" to "I ship code I will never read"
February 19, 2026 at 1:51 PM
amazing how fast we transitioned from "everything should be rewritten in rust for memory safety" to "I ship code I will never read"
it's so tempting with ai to build the superficially functional version first. build the hard bit first. the facade will emerge naturally.
February 19, 2026 at 1:35 PM
it's so tempting with ai to build the superficially functional version first. build the hard bit first. the facade will emerge naturally.