Robert Chandler
@r-cha.dev
Software Engineer, Musician, Follower of Jesus
Everybody's writing posts nobody reads about how they use AI, nobody is writing about their conceptualization of latent spatial navigation (I am) r-cha.dev/blog/how-i-p...
How I Prompt
Intuitively Visualizing and Navigating N-Dimensional Space
r-cha.dev
August 26, 2025 at 6:31 PM
Everybody's writing posts nobody reads about how they use AI, nobody is writing about their conceptualization of latent spatial navigation (I am) r-cha.dev/blog/how-i-p...
The best tool is the one you use
August 7, 2025 at 5:13 PM
The best tool is the one you use
Reposted by Robert Chandler
We’ve heard of p-hacking. Now xAI is p-doom-hacking
Grok’s going to get one of Musk’s updates again. Can’t wait to see what conspiracy theory it just repeats nonstop this time.
June 19, 2025 at 4:46 AM
We’ve heard of p-hacking. Now xAI is p-doom-hacking
ChatGPT when I give it my most mundane idea yet
May 9, 2025 at 2:25 AM
ChatGPT when I give it my most mundane idea yet
Cursor for this, Cursor for that; who’s gonna make Cursor for Programming
May 6, 2025 at 9:24 PM
Cursor for this, Cursor for that; who’s gonna make Cursor for Programming
Who’s making The Social Dilemma II? The AI is ready for you 🔔
April 27, 2025 at 10:00 PM
Who’s making The Social Dilemma II? The AI is ready for you 🔔
So what do I do if my client asked for a simple CSV and made them a whole web app too
April 26, 2025 at 3:25 PM
So what do I do if my client asked for a simple CSV and made them a whole web app too
Turns out, Slack supports HDR for profile pictures and custom emojis. I made a tool to hdr-deepfry your favorite slack emojis and stand out (in the worst way) at work:
hdrify.me
hdrify.me
HDRify
hdrify.me
April 22, 2025 at 9:30 PM
Turns out, Slack supports HDR for profile pictures and custom emojis. I made a tool to hdr-deepfry your favorite slack emojis and stand out (in the worst way) at work:
hdrify.me
hdrify.me
Unlike social media, that later got engineered to exploit us, with AI it's the modality itself + how we naturally engage as humans. It took <1 yr for the “don’t anthropomorphize the LLM’s” voices to be drowned out because we can’t easily reason about anything vaguely resembling humanity (pareidolia)
April 21, 2025 at 4:34 PM
Unlike social media, that later got engineered to exploit us, with AI it's the modality itself + how we naturally engage as humans. It took <1 yr for the “don’t anthropomorphize the LLM’s” voices to be drowned out because we can’t easily reason about anything vaguely resembling humanity (pareidolia)
As you increasingly feel the urge to cast your cares on chatbots, recognize that AI is the next social media. In a decade or two, you’ll be as wireheaded as the boomers slurping Facebook slop *unless you only use AI for things it’s actually good at* i.e. as a tool, not a friend or source of truth.
April 20, 2025 at 1:00 AM
As you increasingly feel the urge to cast your cares on chatbots, recognize that AI is the next social media. In a decade or two, you’ll be as wireheaded as the boomers slurping Facebook slop *unless you only use AI for things it’s actually good at* i.e. as a tool, not a friend or source of truth.
Devs need to invest time and practice into this new generation of tools. Doing so is not a commitment to vibe code; you can still use them thoughtfully to complement your workflows.
www.anthropic.com/engineering/...
www.anthropic.com/engineering/...
Claude Code Best Practices
A blog post covering tips and tricks that have proven effective for using Claude Code across various codebases, languages, and environments.
www.anthropic.com
April 19, 2025 at 6:50 PM
Devs need to invest time and practice into this new generation of tools. Doing so is not a commitment to vibe code; you can still use them thoughtfully to complement your workflows.
www.anthropic.com/engineering/...
www.anthropic.com/engineering/...
It’s software malpractice to force users to move to a new app and it’s half a gigabyte.
March 18, 2025 at 4:16 PM
It’s software malpractice to force users to move to a new app and it’s half a gigabyte.
Reposted by Robert Chandler
I’ve now heard several companies having created their own “job function eval” test. Where eg data analysts put together a list of 100-200 tasks/questions an analyst should solve, near-flawless.
Then feed it into AI models.
And get a sense on where models *really* are!
Then feed it into AI models.
And get a sense on where models *really* are!
March 8, 2025 at 8:01 PM
I’ve now heard several companies having created their own “job function eval” test. Where eg data analysts put together a list of 100-200 tasks/questions an analyst should solve, near-flawless.
Then feed it into AI models.
And get a sense on where models *really* are!
Then feed it into AI models.
And get a sense on where models *really* are!
Wrote down some thoughts on emergent misalignment, more philosophical than technical
r-cha.dev/blog/emergen...
r-cha.dev/blog/emergen...
Emergent Misalignment
A response to recent AI safety research
r-cha.dev
March 7, 2025 at 7:44 PM
Wrote down some thoughts on emergent misalignment, more philosophical than technical
r-cha.dev/blog/emergen...
r-cha.dev/blog/emergen...
I made omegle but for music
tradingfours.app
tradingfours.app
Trading Fours - Make Music with Strangers
A piano interface where strangers connect to trade musical phrases, creating spontaneous collaborations. No login required.
tradingfours.app
March 3, 2025 at 9:19 PM
I made omegle but for music
tradingfours.app
tradingfours.app
The research that showed broad LLM misalignment from a single undesirable behavior has had me thinking. I drew a parallel to human behavior, but I think the conclusions about humanity people draw from LLM behavior say a lot more about humanity than the LLM behavior itself.
February 26, 2025 at 9:07 PM
The research that showed broad LLM misalignment from a single undesirable behavior has had me thinking. I drew a parallel to human behavior, but I think the conclusions about humanity people draw from LLM behavior say a lot more about humanity than the LLM behavior itself.
This probably says a lot more about my code than the state of AI, but Claude Code is the first tool to implement an entire feature in my app from a single prompt. $2 for the session.
Pretty thrilled with the possibilities here.
Pretty thrilled with the possibilities here.
February 24, 2025 at 10:51 PM
This probably says a lot more about my code than the state of AI, but Claude Code is the first tool to implement an entire feature in my app from a single prompt. $2 for the session.
Pretty thrilled with the possibilities here.
Pretty thrilled with the possibilities here.
Reposted by Robert Chandler
Want to use federal data? The U.S. government used to have a prominent link to sign up for API access; that link has been removed, but the form to get it remains operative. You may wish to sign up now, on the premise you might not be able to later. api.data.gov/signup/
February 13, 2025 at 9:02 PM
Want to use federal data? The U.S. government used to have a prominent link to sign up for API access; that link has been removed, but the form to get it remains operative. You may wish to sign up now, on the premise you might not be able to later. api.data.gov/signup/
I think LLM’s should be doing program synthesis in about 90% of cases where labs + co’s are instead reaching for “inference-time compute”
The biggest problem is how tasks where these things "lie" are ones when they need to output a LOT of tokens.
Midway, they stop.
Call it lazy: or perhaps cost efficiency.
You know what doesn't "stop" midway? Regex, traditional programming & non-LLM operations like find+replace.
Midway, they stop.
Call it lazy: or perhaps cost efficiency.
You know what doesn't "stop" midway? Regex, traditional programming & non-LLM operations like find+replace.
February 13, 2025 at 3:13 PM
I think LLM’s should be doing program synthesis in about 90% of cases where labs + co’s are instead reaching for “inference-time compute”
Good news: I just had another idea
Bad news: I just bought another domain
Bad news: I just bought another domain
February 8, 2025 at 7:06 PM
Good news: I just had another idea
Bad news: I just bought another domain
Bad news: I just bought another domain
Nobody:
Me: Yeah so I'm kinda DESTROYING in Advent of Code this year.
Me: Yeah so I'm kinda DESTROYING in Advent of Code this year.
December 19, 2024 at 12:13 AM
Nobody:
Me: Yeah so I'm kinda DESTROYING in Advent of Code this year.
Me: Yeah so I'm kinda DESTROYING in Advent of Code this year.
Nothing like an educational pamphlet on computing from 1957 to give a sense of wonder and awe to your 9-5 writing CRUD apps
October 23, 2024 at 3:48 PM
Nothing like an educational pamphlet on computing from 1957 to give a sense of wonder and awe to your 9-5 writing CRUD apps