Johan Carlin
@johancarlin.com
Python, data, AI. Recovering academic.
twitter.com/johancarlin - abandoned for obvious reasons
https://fosstodon.org/@johancarlin
twitter.com/johancarlin - abandoned for obvious reasons
https://fosstodon.org/@johancarlin
Leading engineers is bad enough. Half the time you feel badly out of your depth. You handwave at a technical approach and hint at the business value and good engineers take that and build something amazing. It's easy to feel like an imposter but I've seen what a difference that direction makes ...
February 6, 2026 at 8:38 PM
Leading engineers is bad enough. Half the time you feel badly out of your depth. You handwave at a technical approach and hint at the business value and good engineers take that and build something amazing. It's easy to feel like an imposter but I've seen what a difference that direction makes ...
Ok great so terminal wins for agents but then the question becomes how do you sandbox it effectively because the current YOLO approach is going to become a security disaster if widely adopted
February 6, 2026 at 12:31 PM
Ok great so terminal wins for agents but then the question becomes how do you sandbox it effectively because the current YOLO approach is going to become a security disaster if widely adopted
Reposted by Johan Carlin
One of the problems with being UTC+1 is that my morning bluesky feed is mostly late night, bleary-eyed, two-drinks-in American expressions of rage and grief over the death of the republic. It's quite the way to wake up
January 28, 2026 at 7:12 AM
One of the problems with being UTC+1 is that my morning bluesky feed is mostly late night, bleary-eyed, two-drinks-in American expressions of rage and grief over the death of the republic. It's quite the way to wake up
Surprised to see OpenAI disable SSL validation here in the chatgpt sandbox. I'm sure it's a manageable risk in their setup but kind of unnecessary to accept it
January 26, 2026 at 7:42 PM
Surprised to see OpenAI disable SSL validation here in the chatgpt sandbox. I'm sure it's a manageable risk in their setup but kind of unnecessary to accept it
Reposted by Johan Carlin
A very sane AI usage policy for any open source project that still cares about quality.
AI Usage Policy
https://github.com/ghostty-org/ghostty/blob/main/AI_POLICY.md
https://news.ycombinator.com/item?id=46730504
https://github.com/ghostty-org/ghostty/blob/main/AI_POLICY.md
https://news.ycombinator.com/item?id=46730504
ghostty/AI_POLICY.md at main · ghostty-org/ghostty
👻 Ghostty is a fast, feature-rich, and cross-platform terminal emulator that uses platform-native UI and GPU acceleration. - ghostty-org/ghostty
github.com
January 23, 2026 at 6:10 PM
A very sane AI usage policy for any open source project that still cares about quality.
Reposted by Johan Carlin
One of the hardest thing for humans to do consistently is get risk factors in the correct order. Confusing primary effects with third order fourth order or fifth order effects is shockingly human.
This kid is now over 20… I wish some reporter would do a follow up story
This kid is now over 20… I wish some reporter would do a follow up story
January 23, 2026 at 2:04 AM
One of the hardest thing for humans to do consistently is get risk factors in the correct order. Confusing primary effects with third order fourth order or fifth order effects is shockingly human.
This kid is now over 20… I wish some reporter would do a follow up story
This kid is now over 20… I wish some reporter would do a follow up story
The Lidl panettone has been open for a month now and it's still completely fresh. A few more Christmases and when it's time I won't need embalming
January 22, 2026 at 6:22 PM
The Lidl panettone has been open for a month now and it's still completely fresh. A few more Christmases and when it's time I won't need embalming
Reposted by Johan Carlin
the rate of change in the AI space is so fast that i cannot explain to my work colleagues that the API we have access to (GPT-4o) is so outdated as to be useless because there's no single chart that depicts both models on the same benchmarks.
January 21, 2026 at 10:16 PM
the rate of change in the AI space is so fast that i cannot explain to my work colleagues that the API we have access to (GPT-4o) is so outdated as to be useless because there's no single chart that depicts both models on the same benchmarks.
Spent stolen moments during the holidays vibing a refresh of my old blog with codex web. I came away with a fancy Sveltia CMS and an appreciation for Codex Web, which is limited yet highly addictive www.johancarlin.com/vibing-my-bl...
Vibing My Blog Back to Life with Codex Web
''
www.johancarlin.com
January 21, 2026 at 7:19 PM
Spent stolen moments during the holidays vibing a refresh of my old blog with codex web. I came away with a fancy Sveltia CMS and an appreciation for Codex Web, which is limited yet highly addictive www.johancarlin.com/vibing-my-bl...
There's a step down in agent performance from software development to standard office apps. In part it's that the tooling is not there yet (office/gdocs gui integrations are crude compared to VS Code).
But in part there's just no feedback loop without automatic tests and that's harder to fix
But in part there's just no feedback loop without automatic tests and that's harder to fix
January 20, 2026 at 6:02 AM
There's a step down in agent performance from software development to standard office apps. In part it's that the tooling is not there yet (office/gdocs gui integrations are crude compared to VS Code).
But in part there's just no feedback loop without automatic tests and that's harder to fix
But in part there's just no feedback loop without automatic tests and that's harder to fix
Reposted by Johan Carlin
Creator of node.js
January 20, 2026 at 3:25 AM
Creator of node.js
Reposted by Johan Carlin
Brian Eno: "The great benefit of computer sequencers is that they remove the issue of skill, and replace it with the issue of judgement."
January 19, 2026 at 3:28 PM
Brian Eno: "The great benefit of computer sequencers is that they remove the issue of skill, and replace it with the issue of judgement."
Spent a bit of time vibing from my phone with codex web. The big showstopper so far is it has no way to handle merge conflicts or indeed respond to any incoming changes from the feature branch. Makes it tricky to run parallel tasks
January 16, 2026 at 6:42 PM
Spent a bit of time vibing from my phone with codex web. The big showstopper so far is it has no way to handle merge conflicts or indeed respond to any incoming changes from the feature branch. Makes it tricky to run parallel tasks
Recruiting developers is tricky these days, half the CVs are LLM-reworded versions of your job ad...
January 15, 2026 at 12:28 PM
Recruiting developers is tricky these days, half the CVs are LLM-reworded versions of your job ad...
Reposted by Johan Carlin
Reposted by Johan Carlin
The future is unevenly distributed.
January 12, 2026 at 9:21 AM
The future is unevenly distributed.
Reposted by Johan Carlin
We really need to come up with better metaphors for AI agents than junior developers and interns. It’s far more of a reflection of how disrespectfully we treat those roles than it is a useful metaphor for how agents behave
January 3, 2026 at 9:20 AM
We really need to come up with better metaphors for AI agents than junior developers and interns. It’s far more of a reflection of how disrespectfully we treat those roles than it is a useful metaphor for how agents behave
Stateful agentic LLMs are cool and all but I hope we get past the anthropomorphizing tamagotchi phase soon. I don't think it's a helpful mental model to think of these things as individuals with agency
January 2, 2026 at 8:35 PM
Stateful agentic LLMs are cool and all but I hope we get past the anthropomorphizing tamagotchi phase soon. I don't think it's a helpful mental model to think of these things as individuals with agency
Reposted by Johan Carlin
Has nobody dealt with their legal, finance, and procurement folks at their companies? Y'all realize that people buy software to have "a throat to choke", right?
The functionality is... not that important. It's why RedHat exists, for example: the software is free, but big companies don't *want* free
The functionality is... not that important. It's why RedHat exists, for example: the software is free, but big companies don't *want* free
I think this is exactly the math that is going to drastically change the small SaaS landscape
why would I pay monthly for <tool> when the same cost (maybe less!) can make that and also anything else I need
why would I pay monthly for <tool> when the same cost (maybe less!) can make that and also anything else I need
I pay $100 a month for Xero (accounting software), and I do wonder if upgrading to Claude Code Max for a month or two would pay for itself to write a robust replacement.
It probably won’t get bank feed imports right, but that’s broken in Xero for some accounts anyway.
It probably won’t get bank feed imports right, but that’s broken in Xero for some accounts anyway.
December 30, 2025 at 9:08 PM
Has nobody dealt with their legal, finance, and procurement folks at their companies? Y'all realize that people buy software to have "a throat to choke", right?
The functionality is... not that important. It's why RedHat exists, for example: the software is free, but big companies don't *want* free
The functionality is... not that important. It's why RedHat exists, for example: the software is free, but big companies don't *want* free
Reposted by Johan Carlin
The idea that code-first BI tools will gain traction so that AI workflows can generate end to end analytics pipelines that work with both the opinions of the tools and the opinions on the developers is really compelling.
open.substack.com/pub/hanson37...
open.substack.com/pub/hanson37...
Why Opinionated, Code-First Systems Are the Future of Data
How AI amplifies strong abstractions instead of replacing them
open.substack.com
December 30, 2025 at 1:09 PM
The idea that code-first BI tools will gain traction so that AI workflows can generate end to end analytics pipelines that work with both the opinions of the tools and the opinions on the developers is really compelling.
open.substack.com/pub/hanson37...
open.substack.com/pub/hanson37...
Reposted by Johan Carlin
background terminal in Codex seems like it would really speed things up
December 29, 2025 at 1:14 AM
background terminal in Codex seems like it would really speed things up
The occasional Google inactive account manager reminder emails are a real memento mori
December 17, 2025 at 5:48 AM
The occasional Google inactive account manager reminder emails are a real memento mori
Having fun playing architect with Gemini. It one-shotted this diagram of a Grafana/Prometheus/Loki logging/monitoring stack for LLMs deployed to k8s with Nvidia NIM. Pretty inspiring
December 16, 2025 at 9:40 AM
Having fun playing architect with Gemini. It one-shotted this diagram of a Grafana/Prometheus/Loki logging/monitoring stack for LLMs deployed to k8s with Nvidia NIM. Pretty inspiring
This place is a bit like vim. Amazing potential but the defaults are poor and you need to be always tinkering with your configuration to have the best experience
December 12, 2025 at 4:05 PM
This place is a bit like vim. Amazing potential but the defaults are poor and you need to be always tinkering with your configuration to have the best experience