Alex Gude
@alexgude.com
Machine learning engineer in Silicon Valley! Formerly a particle physicist at CERN and alumni of Insight. He/Him
https://alexgude.com
@[email protected]
https://alexgude.com
@[email protected]
Pinned
Alex Gude
@alexgude.com
· Nov 26
The Darfsteller
The Darfsteller, by Walter M. Miller Jr., is a Hugo Award-winning novelette about the obsolescence of the human artist. It follows Ryan Thornier, a former stage idol reduced to working as a janitor in...
alexgude.com
Seventy years ago Walter M Miller Jr wrote how creative work would be the first thing we automate. The Darfsteller follows an actor replaced by robots, which are both cheaper and preferred by the audience.
It's more relevant than ever in the age of Gen AI.
alexgude.com/books/the_da...
#BookReview
It's more relevant than ever in the age of Gen AI.
alexgude.com/books/the_da...
#BookReview
So I was doubly wrong!
First, tea is only up $16 -> $17 in the year (although I guess we have 3 months to go, and who knows how WW3 will impact prices).
And second, it wasn't a year's supply!
First, tea is only up $16 -> $17 in the year (although I guess we have 3 months to go, and who knows how WW3 will impact prices).
And second, it wasn't a year's supply!
This post is how I remembered to go order a year's worth of tea before the price doubles. 🙃
January 21, 2026 at 5:52 AM
So I was doubly wrong!
First, tea is only up $16 -> $17 in the year (although I guess we have 3 months to go, and who knows how WW3 will impact prices).
And second, it wasn't a year's supply!
First, tea is only up $16 -> $17 in the year (although I guess we have 3 months to go, and who knows how WW3 will impact prices).
And second, it wasn't a year's supply!
I really should write up how Hamilton's Pandora's Star and Judas Unchained are a surface level copy of Simmons's Hyperion Cantos... 🤔
January 21, 2026 at 3:16 AM
I really should write up how Hamilton's Pandora's Star and Judas Unchained are a surface level copy of Simmons's Hyperion Cantos... 🤔
Reposted by Alex Gude
no greenlander ever called me a libtard
January 20, 2026 at 8:18 PM
no greenlander ever called me a libtard
Reposted by Alex Gude
This was 5 months ago. Things have gotten substantially more powerful since then.
I just cranked out 120k of full stack code in roughly a sprint. That required me stopping dozens of prompts mid cycle because I saw that the LLM was fucking something up.
If you're a junior who never learns to detect that stuff.. what does that look like in 5-10 years?
If you're a junior who never learns to detect that stuff.. what does that look like in 5-10 years?
January 20, 2026 at 4:46 AM
This was 5 months ago. Things have gotten substantially more powerful since then.
Reposted by Alex Gude
Creator of node.js
January 20, 2026 at 3:25 AM
Creator of node.js
Reposted by Alex Gude
This is a perfect piece of technical writing. alexharri.com/blog/ascii-r...
ASCII characters are not pixels: a deep dive into ASCII rendering
A look at how I used shape vectors to achieve sharp, high-quality ASCII rendering.
alexharri.com
January 17, 2026 at 6:07 PM
This is a perfect piece of technical writing. alexharri.com/blog/ascii-r...
The "This is the most fun I've had..." sentiment follows my experience as well.
I'll add: it's the most powerful computers have felt to me since I first learned to code. Now it's easier than ever to make them do what I want them to.
I'll add: it's the most powerful computers have felt to me since I first learned to code. Now it's easier than ever to make them do what I want them to.
After making dozens of prototypes with AI coding agents, here's what I think: human developers are still essential
In fact, instead of taking jobs, this new class of software tools may make workers busier than ever. I wrote about my experience for Ars Technica:
arstechnica.com/information-...
In fact, instead of taking jobs, this new class of software tools may make workers busier than ever. I wrote about my experience for Ars Technica:
arstechnica.com/information-...
10 things I learned from burning myself out with AI coding agents
Opinion: As software power tools, AI agents may make people busier than ever before.
arstechnica.com
January 19, 2026 at 2:54 PM
The "This is the most fun I've had..." sentiment follows my experience as well.
I'll add: it's the most powerful computers have felt to me since I first learned to code. Now it's easier than ever to make them do what I want them to.
I'll add: it's the most powerful computers have felt to me since I first learned to code. Now it's easier than ever to make them do what I want them to.
Reposted by Alex Gude
After making dozens of prototypes with AI coding agents, here's what I think: human developers are still essential
In fact, instead of taking jobs, this new class of software tools may make workers busier than ever. I wrote about my experience for Ars Technica:
arstechnica.com/information-...
In fact, instead of taking jobs, this new class of software tools may make workers busier than ever. I wrote about my experience for Ars Technica:
arstechnica.com/information-...
10 things I learned from burning myself out with AI coding agents
Opinion: As software power tools, AI agents may make people busier than ever before.
arstechnica.com
January 19, 2026 at 2:08 PM
After making dozens of prototypes with AI coding agents, here's what I think: human developers are still essential
In fact, instead of taking jobs, this new class of software tools may make workers busier than ever. I wrote about my experience for Ars Technica:
arstechnica.com/information-...
In fact, instead of taking jobs, this new class of software tools may make workers busier than ever. I wrote about my experience for Ars Technica:
arstechnica.com/information-...
Oh, so Endymion is just Terminator 2 in Hyperion-verse eh?
January 19, 2026 at 5:40 AM
Oh, so Endymion is just Terminator 2 in Hyperion-verse eh?
Reposted by Alex Gude
i don't think there is any desk job i have ever done that i could not have done better by having Claude Code to hand. being able to build bespoke widgets at just below the speed of language is really, really good!
these tools were not good at dev tasks until a massive amount of resources made them fantastic at them. i do believe other fields will be more difficult.
but if you are counting these tools out and have confidence they'll never be able to work.. i do not understand that confidence.
but if you are counting these tools out and have confidence they'll never be able to work.. i do not understand that confidence.
January 18, 2026 at 11:03 PM
i don't think there is any desk job i have ever done that i could not have done better by having Claude Code to hand. being able to build bespoke widgets at just below the speed of language is really, really good!
Friday has a high of -10 °F?
Nice!
Nice!
January 18, 2026 at 10:57 PM
Friday has a high of -10 °F?
Nice!
Nice!
Related:
Antigravity has some good ideas built into it like commenting on a plan before implementing it.
But boy is it bad at exploring the repo before it gets started. 😑
Antigravity has some good ideas built into it like commenting on a plan before implementing it.
But boy is it bad at exploring the repo before it gets started. 😑
With coding agents theres just no real competition for anthropic right now, which is...not ideal! Claude code is *good*, its really good but nobody else comes close right now despite what the benchmarks say (benchmarks are bullshit). Someone else needs to get on their level!
January 18, 2026 at 9:48 PM
Related:
Antigravity has some good ideas built into it like commenting on a plan before implementing it.
But boy is it bad at exploring the repo before it gets started. 😑
Antigravity has some good ideas built into it like commenting on a plan before implementing it.
But boy is it bad at exploring the repo before it gets started. 😑
Reposted by Alex Gude
Erdos problems, a set of famous difficult math challenges, are a clear example of AI models breaching a threshold. The idea that an AI could solve one, let alone many, would have been insane a year ago (o1 was brand new). Now we have multiple Erdos problems solved by GPT-5.2 in the last couple weeks
January 18, 2026 at 3:38 AM
Erdos problems, a set of famous difficult math challenges, are a clear example of AI models breaching a threshold. The idea that an AI could solve one, let alone many, would have been insane a year ago (o1 was brand new). Now we have multiple Erdos problems solved by GPT-5.2 in the last couple weeks
Reposted by Alex Gude
If you're a developer surrounded by people debating the utility of LLMs, you are being misled.
The advent of agents decisively settles the conversation. The real question is: how do you ship software that leverages their power?
I spent 2025 answering this question
networkgames.fyi/the-year-eve...
The advent of agents decisively settles the conversation. The real question is: how do you ship software that leverages their power?
I spent 2025 answering this question
networkgames.fyi/the-year-eve...
The year everything changed
Far from my best years being behind me, it feels as though everything in my life has been leading to this moment: building a completely new category of product where the most useful lessons come not f...
networkgames.fyi
January 17, 2026 at 4:52 PM
If you're a developer surrounded by people debating the utility of LLMs, you are being misled.
The advent of agents decisively settles the conversation. The real question is: how do you ship software that leverages their power?
I spent 2025 answering this question
networkgames.fyi/the-year-eve...
The advent of agents decisively settles the conversation. The real question is: how do you ship software that leverages their power?
I spent 2025 answering this question
networkgames.fyi/the-year-eve...
Reposted by Alex Gude
I absolutely love this line and I am going to say it, verbatim, to every other developer during our next company-wide meeting.
long before Claude Code writes its first line of code for you, it earns its value by being the most effective diagnostic and debugging tool in computing history
this is nowhere in the conversation. don't get your time wasted
this is nowhere in the conversation. don't get your time wasted
January 17, 2026 at 11:02 PM
I absolutely love this line and I am going to say it, verbatim, to every other developer during our next company-wide meeting.
Reposted by Alex Gude
With “Wikimedia Enterprise”, AI companies have to use (and pay for) dedicated APIs to scrape data, which helps to limit the strain on Wikimedia servers. (See eg arstechnica.com/information-...) This is a good thing for Wikimedia and for its readers.
AI bots strain Wikimedia as bandwidth surges 50%
Automated AI bots seeking training data threaten Wikipedia project stability, foundation says.
arstechnica.com
January 15, 2026 at 6:52 PM
With “Wikimedia Enterprise”, AI companies have to use (and pay for) dedicated APIs to scrape data, which helps to limit the strain on Wikimedia servers. (See eg arstechnica.com/information-...) This is a good thing for Wikimedia and for its readers.
Reposted by Alex Gude
In google meet you can now react with any emoji instead of the usual short list (clap, heart, etc)
*Importantly* if you use the crab emoji it scuttles horizontally across the screen, and this has a choke-hold on our larger meetings 🦀 especially if Rust was mentioned
*Importantly* if you use the crab emoji it scuttles horizontally across the screen, and this has a choke-hold on our larger meetings 🦀 especially if Rust was mentioned
January 16, 2026 at 4:39 PM
In google meet you can now react with any emoji instead of the usual short list (clap, heart, etc)
*Importantly* if you use the crab emoji it scuttles horizontally across the screen, and this has a choke-hold on our larger meetings 🦀 especially if Rust was mentioned
*Importantly* if you use the crab emoji it scuttles horizontally across the screen, and this has a choke-hold on our larger meetings 🦀 especially if Rust was mentioned
Reposted by Alex Gude
Feeling increasingly over arguing with antis about AI.
The last few months prove that we're in a takeoff situation. It's not a FOOM situation, but we essentially have AGI at this point. The newest models are truly amazing. The cynics' criticisms are entirely obsolete these days.
The last few months prove that we're in a takeoff situation. It's not a FOOM situation, but we essentially have AGI at this point. The newest models are truly amazing. The cynics' criticisms are entirely obsolete these days.
January 16, 2026 at 2:32 AM
Feeling increasingly over arguing with antis about AI.
The last few months prove that we're in a takeoff situation. It's not a FOOM situation, but we essentially have AGI at this point. The newest models are truly amazing. The cynics' criticisms are entirely obsolete these days.
The last few months prove that we're in a takeoff situation. It's not a FOOM situation, but we essentially have AGI at this point. The newest models are truly amazing. The cynics' criticisms are entirely obsolete these days.
Hey chat, is it good that our NATO allies are deploying tripwire forces to deter us?
SWEDEN PRIME MINISTER: SOME OFFICERS FROM THE SWEDISH ARMED FORCES ARRIVE IN GREENLAND TODAY
January 14, 2026 at 4:39 PM
Hey chat, is it good that our NATO allies are deploying tripwire forces to deter us?
I'm running 6 Claude Code sessions over three projects right now and it's the first thing that's made me go "You know, I think I need 4 monitors..."
January 13, 2026 at 7:15 PM
I'm running 6 Claude Code sessions over three projects right now and it's the first thing that's made me go "You know, I think I need 4 monitors..."
I've been thinking about how LLMs produce code so fast that you can't human review it all.
That means we either need to slow down, or make checking easier.
Things like type systems, compilers, tests, etc. are probably part of the solution there.
So anyway, I guess I should listen to the podcast.
That means we either need to slow down, or make checking easier.
Things like type systems, compilers, tests, etc. are probably part of the solution there.
So anyway, I guess I should listen to the podcast.
strong types make it far easier for the tools, not harder. type errors are a very clear signal! check out this podcast in a few days
Today, @ahl.bsky.social and I are going to take on a hot topic: what does engineering rigor look like in the age of LLMs? We will be joined by Oxide engineers who have been at the forefront of using LLMs to increase the rigor in their own work. Join us, 5p Pacific!
discord.gg/QrcKGTTPrF?e...
discord.gg/QrcKGTTPrF?e...
January 13, 2026 at 7:11 PM
I've been thinking about how LLMs produce code so fast that you can't human review it all.
That means we either need to slow down, or make checking easier.
Things like type systems, compilers, tests, etc. are probably part of the solution there.
So anyway, I guess I should listen to the podcast.
That means we either need to slow down, or make checking easier.
Things like type systems, compilers, tests, etc. are probably part of the solution there.
So anyway, I guess I should listen to the podcast.
Reposted by Alex Gude
stonks are flat due to the jpow firing thing because if he doesn't get put in jail the economy is in good hands and if he does get put in in jail you'd rather be in stocks for the hyperinflation
January 13, 2026 at 4:11 PM
stonks are flat due to the jpow firing thing because if he doesn't get put in jail the economy is in good hands and if he does get put in in jail you'd rather be in stocks for the hyperinflation
Believe it or not this is the plot of Iain M. Banks's _Matter_.
Umm... spoilers I guess!
Umm... spoilers I guess!
the hungry ghost
it destroyed its jar
yes
YES
the ghost is out
it destroyed its jar
yes
YES
the ghost is out
January 13, 2026 at 12:00 AM
Believe it or not this is the plot of Iain M. Banks's _Matter_.
Umm... spoilers I guess!
Umm... spoilers I guess!
Reposted by Alex Gude
(Finally my time to shine!)
They're not, Boy and Girl popular names behave differently:
alexgude.com/blog/popular...
They're not, Boy and Girl popular names behave differently:
alexgude.com/blog/popular...
Rise and Fall of Popular Names
The popularity of baby names rises and falls based on the tastes of each generation of parents. Are their preferences the same for boy’s names as for girl’s names? I plot the trends to find out!
alexgude.com
January 12, 2026 at 11:13 PM
(Finally my time to shine!)
They're not, Boy and Girl popular names behave differently:
alexgude.com/blog/popular...
They're not, Boy and Girl popular names behave differently:
alexgude.com/blog/popular...
Reposted by Alex Gude
The future is unevenly distributed.
January 12, 2026 at 9:21 AM
The future is unevenly distributed.