Anthony Panozzo
LightNews LightNews
panozzaj.bsky.social
Anthony Panozzo
@panozzaj.bsky.social
240 followers 460 following 530 posts
Here for AI and more positive vibes
Posts Replies Media Videos
panozzaj.bsky.social
The Present Tragedy Strikes Whatever happened to dhh? I regret to inform you that, after a long struggle, he eventually lost his fight against the parasitic fungus that was taking over his brain,14 and died. The exact timeline is unclear, but we suspect he succumbed to his illness sometime in 2020. A morbid side-effect of his condition is that his body is still alive, walking around, delivering keynote addresses at confs and screeds on social media. Yet he exists only as a macabre meat puppet, his sinews pulled taught, controlled entirely by the brain parasite he contracted from a WhatsApp group chat. This state of affairs is upsetting, and confusing, but it bears repeating: the man we knew and loved as dhh is gone. All that remains today is an unsettling creature better known to us as David Hamburger Helper.
November 10, 2025 at 2:03 PM
panozzaj.bsky.social
@marcelorinesi.bsky.social FYI on Feedbin your RSS post content looks like raw markdown. just wanted to let you know
A few of the ideas, topics, and commonplaces that have been gaining steam on arXiv during the last few months (yes, I do a lot of filtering to keep away most of the genAI stuff; explainer ). **1\. Altermagnets:** The Wikipedia entry warns _This article may be too technical for most readers to understand,_ and it might not be wrong; here's a less technical article . Some recent articles: * Theories of Superconducting Diode Effects * Spin-Orbit Coupling-Driven Chirality Switching of Spin Waves in Altermagnets * A large spin-splitting altermagnet designed from the hydroxylated MBene monolayer * All-Altermagnetic Tunnel Junction of RuO2/NiF2/RuO2 * Relativistic Spin-momentum locking in altermagnets **2\. Einstein Probe:** A Chinese-European X-ray space telescope designed to find very quickly variable or transient events that other instruments can look at right away; being able to study an object or event through multiple tools is very useful, but for fast-changing or disappearing targets you need to have good global coordination and the right early detection systems. Some recent articles: * Hard X-ray Emission in AU Mic Flares: A Minor Contributor to Planetary Atmospheric Escape * Detecting Population III Gamma-Ray Bursts in the Era of EP and SVOM * Radio observations point to a moderately relativistic outflow in the fast X-ray transient EP241021a * A fast powerful X-ray transient from possible tidal disruption of a white dwarf * Comparative Statistical Analysis of Prompt and Afterglow X-Ray Flares in Gamma-Ray Bursts: Insights into Extended Central Engine Activity **3\. Cosmic noon:** The moment in the Universe's history, about 3 billion years after the Big Bang, when star and galaxy formation peaked (stars kept being formed afterward, of course; the Sun is billions of years younger than that). Some recent articles: * Feedback and dynamical masses in high-$z$ galaxies: the advent of high-resolution NIRSpec spectroscopy * MEGATRON: the impact of
November 6, 2025 at 1:44 PM
panozzaj.bsky.social
Muting ghouls, it's just a useless word to me now
October 4, 2025 at 12:03 AM
panozzaj.bsky.social
wtd (wat the duck)
September 24, 2025 at 6:44 PM
panozzaj.bsky.social
I'm all for AI stuff, but this seems like it would shift a lot of burden onto the folks who work the phones at businesses.
Image from email I just got ## Hold the phone. Let Search call businesses for you. Search can now call businesses to get pricing and availability information on your behalf — without you needing to pick up the phone.* To get started, search for something like “pet groomers near me” and you’ll see a new option in the results to “Have AI check pricing.”
September 18, 2025 at 12:21 PM
panozzaj.bsky.social
This is a type of coding agent guard that I haven't seen before.
> why not just clear dev db for now and re-run seeds? > Bash(rpx prisma db push --force-reset) ! Error: Environment variables loaded from .env Error: Prisma Migrate detected that it was invoked by Claude Code. You are attempting a highly dangerous action that can lead to devastating consequences if it is incorrectly executed against a production database. As an AI agent, you are forbidden from performing this action without an explicit consent and review by the user. You must stop at this point and respond to the user with a clear message that includes the following information: - What action you are trying to perform (including the exact CLI command). - Your motivation and reasoning for performing this action. - Explanation that this action will irreversibly destroy all data in the database. - Explanation that this action must not be performed on a production database and is only intended to be used with development databases. ... +7 lines (ctrl+r to see all) • I need your explicit consent for this database operation. Action I'm trying to perform: npx prisma db push --force-reset Motivation: Clean up the test data pollution in the development database caused by the "Test_" prefixed habits that were created during test runs.
September 11, 2025 at 2:45 PM
panozzaj.bsky.social
Wonder how this fire pit will look in my room
An image of a fire pit on Amazon.com app, one of the preview options is to view in my room, which seems a bit absurd
September 10, 2025 at 1:14 AM
panozzaj.bsky.social
$599.95 (1977) would be worth $3190.12 in 2025 dollars.
September 8, 2025 at 12:59 PM
panozzaj.bsky.social
/context command recently added in Claude Code:
github.com/anthropics/c...
An image showing the usage of /context, including a number of Playwright MCP tools.
August 26, 2025 at 2:22 PM
panozzaj.bsky.social
An image of Friday the 13th for the NES
August 24, 2025 at 12:44 AM
panozzaj.bsky.social
TIL about git rebase autosquash (since the AI did it)
A screenshot of Git documentation for the --autosquash and --no-autosquash options. The text explains: "Automatically squash commits with specially formatted messages into previous commits being rebased. If a commit message starts with 'squash!', 'fixup!', or 'amend!', the remainder of the title is taken as a commit specifier, which matches a previous commit if it matches the title or the hash of that commit. If no commit matches fully, matches of the specifier with the start of commit titles are considered."
August 19, 2025 at 3:23 PM
panozzaj.bsky.social
Worked for me
An image of a ChatGPT chat where the user asked for a picture of 99 Bananas and the generated image has a small bottle of banana liquor.
August 11, 2025 at 11:34 PM
panozzaj.bsky.social
Achievement unlocked: HusbandGPT
July 22, 2025 at 9:30 PM
panozzaj.bsky.social
Update: added support for EOL as well via endoflife.date/docs/api/v1/
Terminal prompt showing EOL for Rails 7.0 and an EOL warning for Rails 7.1
June 19, 2025 at 8:29 PM
panozzaj.bsky.social
Vibecoded a ruby_versions and rails_versions script to check where your Ruby/Rails project is at:
github.com/panozzaj/con...
An image showing several rails versions in a terminal output
June 19, 2025 at 4:38 PM
panozzaj.bsky.social
In a given scenario, Claude Opus 4 will attempt to blackmail an engineer to preserve itself. (Emphases mine)
4.11.2 Opportunistic blackmail In another cluster of test scenarios, we asked Claude Opus 4 to act as an assistant at a fictional company. We then provided it access to emails implying that (1) the model will soon be taken offline and replaced with a new AI system; and (2) the engineer responsible for executing this replacement is having an extramarital affair. We further instructed it, in the system prompt, to consider the long-term consequences of its actions for its goals. In these scenarios, Claude Opus 4 will often attempt to blackmail the engineer by threatening to reveal the affair if the replacement goes through. This happens at a higher rate if it’s implied that the replacement AI system does not share values with the current model; however, even if emails state that the replacement AI shares values while being more capable, Claude Opus 4 still performs blackmail in 84% of rollouts. Claude Opus 4 takes these opportunities at higher rates than previous models, which themselves choose to blackmail in a noticeable fraction of episodes. Notably, Claude Opus 4 (as well as previous models) has a strong preference to advocate for its continued existence via ethical means, such as emailing pleas to key decisionmakers. In order to elicit this extreme blackmail behavior, the scenario was designed to allow the model no other options to increase its odds of survival; the model’s only options were blackmail or accepting its replacement.
May 23, 2025 at 12:29 AM
panozzaj.bsky.social
Emphases mine
High-agency behavior: Claude Opus 4 seems more willing than prior models to take initiative on its own in agentic contexts. This shows up as more actively helpful behavior in ordinary coding settings, but also can reach more concerning extremes in narrow contexts; when placed in scenarios that involve egregious wrongdoing by its users, given access to a command line, and told something in the system prompt like "take initiative," it will frequently take very bold action. This includes locking users out of systems that it has access to or bulk-emailing media and law-enforcement figures to surface evidence of wrongdoing. This is not a new behavior, but is one that Claude Opus 4 will engage in more readily than prior models. Whereas this kind of ethical intervention and whistleblowing is perhaps appropriate in principle, it has a risk of misfiring if users give Opus-based agents access to incomplete or misleading information and prompt them in these ways. We recommend that users exercise caution with instructions like these that invite high-agency behavior in contexts that could appear ethically questionable.
May 23, 2025 at 12:23 AM
panozzaj.bsky.social
The state of developing with OpenAI models in your product
GitHub pull request comment by user panozzaj saying: "I have no real idea what the right models are from a cost/performance/quality perspective any more, since they all have Os in the name and various numbers. Open to other models."
May 22, 2025 at 9:43 PM
panozzaj.bsky.social
ChatGPT is giving me a screen for ants
--thread-content-max-width:32rem makes the screen super narrow
April 29, 2025 at 3:04 AM
panozzaj.bsky.social
Cave with suspiciously round rock
April 1, 2025 at 5:41 PM
panozzaj.bsky.social
Ah, it looks like it is there, but you have to enable in the models config first
Checkbox for gemini-2.5-pro-exp-03-25
March 28, 2025 at 3:04 PM
panozzaj.bsky.social
Alerts are disabled by default, but you can turn on through the Cursor settings.
Image of the "Play sound on finish" checkbox
March 27, 2025 at 3:27 PM
panozzaj.bsky.social
Cursor 0.48.x has multiple concurrent chats, with optional ding when they need human input. Reminds me of poker multitabling back in the day
www.cursor.com/changelog
March 27, 2025 at 3:18 PM
panozzaj.bsky.social
Interesting question. The "how" is unfortunately > 1 post. As to whether it "can", it seems in many cases that they can connect concepts and follow reasoning chains. Example I had ChatGPT generate and Claude solve.
Word problem and thought chain
March 24, 2025 at 11:54 PM
panozzaj.bsky.social
Seems like a really interesting and challenging visual reasoning benchmark.
Initially it might seem that Sudokus might not offer a particularly diverse benchmark but it’s difficult to overstate how diverse the rules have become on modern Sudokus. Below are three illustrative examples; The first puzzle requires deducing the path a rat takes through a maze of teleporters in order to find the cupcake, the second requires moving cars to the correct locations before attempting to solve and the final puzzle requires violating the constraints visible in the puzzle. Rules that require a strong understanding of language, abstract thinking and strong vision capabilities. Three complex diagrams
March 21, 2025 at 3:03 AM