iiinova — where your digital self keeps evolving

Reposted by Jan Ehrhardt

Gergely Orosz @gergely.pragmaticengineer.com · 23/09/2026

I am a sucker for good books, and so especially O'Reilly books. Here's one that is WIP, but in return, it is free, thanks to @antithesis.com It's for CTOs / eng leaders who want to get AI to deliver *business value* which is... tricky. Get it here: pages.antithesis.com/oreilly-scal...

2577

Reposted by Jan Ehrhardt

Gus @gusthema.bsky.social · 23/09/2026

I'm glad that the slowdown of AI development finished today with the release of Opus 5.5 and GPT-6 Sol and Luna! Now that all the safety issues are fixed, we can continue to move forward!

0405

Jan Ehrhardt @jehrhardt.dev · 23/09/2026

@oppi.li @anirudh.fi I wonder, if this is an issue for @tangled.org as well.

100

Jan Ehrhardt @jehrhardt.dev · 23/09/2026

I probably don't really need a provider for this, but a tool to run it myself easily would be nice.

100

Jan Ehrhardt @jehrhardt.dev · 23/09/2026

Serverless hosting of coding agent sessions and web apps. I decided that using EU regions on fly.io might be good enough to get started. When demand is proven, I can start renting bare metal machines from Scaleway and run Firecracker myself.

100

Reposted by Jan Ehrhardt

Unsloth AI @unsloth.ai · 22/09/2026

Qwen-Image-2.1 can now run locally on 12GB VRAM with Unsloth GGUFs! The 7B parameter model performs on par with Nano Banana 2.0. Run our Dynamic FP8 or GGUFs for higher quality via diffusers, Unsloth Desktop & more. GGUF: huggingface.co/unsloth/Qwen... Guide: unsloth.ai/docs/models/...

1407

Jan Ehrhardt @jehrhardt.dev · 22/09/2026

I was thinking the same the other day.

130

Reposted by Jan Ehrhardt

Hendrik Mans @hmans.dev · 21/09/2026

I love Workflowy but I also hate that it's the only good true outliner out there and that there's no local-first or local-only option/alternative. I'm already doing too much :( Anyone want to pick this up and vibecode something cool? - Electron/Tauri - Local first - Optional sync via whatever

10192

Jan Ehrhardt @jehrhardt.dev · 21/09/2026

Quitting social media is definitely the better strategy.

001

Reposted by Jan Ehrhardt

conputer dipshit @davidcrespo.bsky.social · 19/09/2026

now this is fun. guest post on Terry Tao's blog by the great Grant Sanderson of 3Blue1Brown arguing that we should "give novel and compelling motivated explanations academic credit similar to what generating new proofs of open problems has had historically."

terrytao.wordpress.com

If math is more than proof, we need to better celebrate the rest of it

[This is a guest post by Grant Sanderson. This blog post was initially written in a different file format and converted using AI. — T.] A sentiment echoing throughout the mathematics communit…

519031

Reposted by Jan Ehrhardt

Matt Henderson @matthen.com · 20/09/2026

How to make an ellipse, using sand and physics

1538648

Reposted by Jan Ehrhardt

Erlend Sogge Heggen @erlend.sh · 19/09/2026

Excellent piece by @seldo.com I’ve noted some connections on Semble: semble.so/url?id=https... imo protective, ‘fair source’ licensing applied sparingly is perfectly compatible with commercialized registries and helps draw clearer boundaries where there currently are none.

2103

Reposted by Jan Ehrhardt

Adina Yakup @adinayakup.bsky.social · 19/09/2026

HyperFlow⚡New MiniMax-H3 variant from Video Rebirth huggingface.co/videorebirth... - Open weight 8 step LoRA - 3× faster with data free self distillation - Video + stereo audio intact

061

Reposted by Jan Ehrhardt

Jonathon @jonathon.lol · 18/09/2026

this is graph of our commits to a single repo going back years with the same team...

618925

Reposted by Jan Ehrhardt

Martin Fowler @martinfowler.com · 17/09/2026

NEW POST I see AI and LLMs as bringing a mix of boons and banes, but one feeling dominates: not just are they pretending to be a person, but the kind of person I don't like. martinfowler.com/articles/202...

Reposted by Jan Ehrhardt

mr. TIM @timkellogg.me · 18/09/2026

Two weeks after the Huggingface incident, attackers broke into OpenAI’s internal systems using Opus 5 and gained write access to their Git repos www.wsj.com/tech/ai/hack...

wsj.com

Exclusive | Hackers Used Anthropic’s Claude to Break Into OpenAI

A bug-hunting independent security research team was able to access OpenAI’s internal code system, exposing growing risks in automated cyber threats.

1110929

Reposted by Jan Ehrhardt

Rust Language @rust-lang.org · 17/09/2026

⚠️ We believe that there is an ongoing campaign targeting owners of popular crates and rust-lang team members that is attempting to compromise devices and accounts in order to use them to publish malware. See our blog post for details: blog.rust-lang.org/2026/09/17/t...

blog.rust-lang.org

Be alert: targeted attacks on prominent Rustaceans | Rust Blog

Empowering everyone to build reliable and efficient software.

121990

Reposted by Jan Ehrhardt

Henry @henryfellerhoff.com · 17/09/2026

Ink & Switch's approach of being forward-looking while staying uncompromising on both the value of intention in what you make and the necessity of shaping the world to match your and your community'´s system of values is (in my opinion) the most essential message in software right now

A quote from Peter van Hardenberg, the lab director of Ink & Switch, from his talk "How to Research Tools for Thought":

"I would encourage everybody to think about how – as we're moving through this new world – that we are intentional about how we shape it. And I don't want to tell you whether you should or shouldn't adopt any particular technology, but I will tell you that the world will not wait for us – and that we need to figure out how we can engage with our values and our intentions, and how we can serve our communities to make sure that the people who are using these tools that we build and that other people build are able to do them in safe and trustworthy ways, but also in ways that are satisfying and joyful and let us have what we set out to do."

0265

Jan Ehrhardt @jehrhardt.dev · 17/09/2026

Btw. did anyone try #jj-vcs to put the spec in the description first and let an agent pick it up from there? Does this even make sense 🤔?

000

Reposted by Jan Ehrhardt

Ky @ky.fyi · 16/09/2026

Meta and X have been quietly downranking external links for ages, but we're now entering a new era of open hostility toward the open web. It's a great time to build your own website, free from any billionaires demanding a tithe for things that have always been free.

5283128

Reposted by Jan Ehrhardt

Gergely Orosz @gergely.pragmaticengineer.com · 16/09/2026

Fresh data from GitHub: Agent-generated PRs have exploded in size. 9x the last 8 months (!!) No signs of slowing down. This is why everyone is rethinking code reviews, deploys, possibly even o11y thanks to the GH team for sharing this data with me! (Slide from my talk yesterday)

65613

Reposted by Jan Ehrhardt

Cult.Repo @cultrepo.bsky.social · 16/09/2026

On Oct 14th 1996, Matthias Ettrich, a student in Germany, posted to Usenet asking for help build a real desktop for Linux. That post led to the creation of KDE, one of the most impactful free software communities in the world. Catch the trailer for our documentary about KDE on Saturday 19th.

28120

Reposted by Jan Ehrhardt

European Space Agency @esa.int · 16/09/2026

Mars isn't always red. 💜 New images from Mars Express reveal striking purple swirls near the planet's south pole. 🔗 esa.int/Science_Exploration/Space_Science/Mars_Express/Purple_swirls_on_the_Red_Planet @science.esa.int

1242889

Reposted by Jan Ehrhardt

Ivy Reese @spiralganglion.com · 15/09/2026

here's a peak at what i've been doing for the past few years: building a calm, playful little space for thinking, doodling, journalling, and programming with materials like ink, paper, and… flux that spreads through your notes like a fluid wait what? www.youtube.com/watch?v=GurW...

youtube.com

PlayBook: A Programmable Paper Notebook

YouTube video by Ink & Switch

35118

Reposted by Jan Ehrhardt

Daily Chartbook @dailychartbook.bsky.social · 14/09/2026

"Weekly tokens on just OpenRouter doubled in just 4 weeks, and as of the last week in August, were up 25x for the year." -Moses Sternstein, Random Walk

0456

Reposted by Jan Ehrhardt

🌱️ @crumb.bsky.social · 14/09/2026

blog post from deepseek kernel engineer mp.weixin.qq.com/s/zk0KxuLzhm...

727042

Reposted by Jan Ehrhardt

Alex Warth @alexwarth.bsky.social · 14/09/2026

Nearly every general-purpose programming language is Turing complete. So ask not what you can do with a language, but what a language makes worth the effort. A short note on "effective expressiveness": www.inkandswitch.com/malleable-so...

inkandswitch.com

Effective expressiveness

Everything is possible. What’s actually worth the effort?

0327

Reposted by Jan Ehrhardt

Jay 🦋 @jay.bsky.team · 15/09/2026

when facing an unstoppable force, focus on changing its course.

2734130

Reposted by Jan Ehrhardt

Simon Willison @simonwillison.net · 14/09/2026

Built a little local web app to help edit commit messages for a repo simonwillison.net/2026/Sep/14/...

Screenshot of the commit-rewriter web interface. A heading reads commit-rewriter above the repository path and current branch and commit hash, with a short description of the tool. A toolbar shows a pending edits count with Discard drafts and Rewrite commit messages buttons, followed by a search box for message, author, or hash and an Edited only checkbox. A left sidebar titled Navigate commits lists recent commit messages with their short hashes. The main panel shows a card for each commit with its hash, author and timestamp, an editable text area containing the commit message, and a View full formatted diff toggle.

7594

Reposted by Jan Ehrhardt

Nathan Lambert @natolambert.bsky.social · 13/09/2026

Open-Source AI & Open Models Reading List How to get up to speed on open models and their implications. www.interconnects.ai/p/open-sourc...

interconnects.ai

Open-Source AI & Open Models Reading List

How to get up to speed on open models and their implications.

1414

Reposted by Jan Ehrhardt

samhain ⎔ @personhood.removal.surgery · 12/09/2026

ctrl+F "china", yup, it's the usual dario dementia, close tab. nothing of value there. he's not urging a "slowdown", he's spinning a story to lure you into reading his sinophobic manifesto about how everyone needs to "keep [US]’ AI lead over [China] as large as possible" or the world ends.

$Pacing within democracies will be limited by the lead that US companies have over authoritarian regimes, chiefly the Chinese Communist Party. If we slow down by more than this amount, then (unpaced) CCP-associated projects will pull ahead, creating significant national security risk. I agree with Secretary Bessent that a Chinese lead in AI would pose grave danger for the United States and the world. The CCP-associated projects will run the alignment risks that US companies are carefully preventing, and even if they avoid those risks, they will be in a position to militarily dominate democracies (for example with AI-driven drones). Thus, a key part of pacing within democracies is to keep democracies’ AI lead over autocracies as large as possible, to give us the breathing room we need in order to pace effectively. The main steps we can take to defend this gap are: Do not sell powerful AI chips or semiconductor manufacturing equipment to China, and crack down on chip smuggling operations and remote access to data centers outside China. Chips will be the main determinant of China’s AI strength. Crack down on unauthorized distillation by companies in authoritarian countries. Distillation of frontier models allows lagging companies to narrow the gap using a fraction of the cost it would take to develop their own AI independently. Strengthen security at the AI companies and prevent model weight theft. Companies and the US government should cooperate to make these steps as effective as possible. Anthropic has consistently advocated for all of these measures, because we’ve always understood that they would be essential to any pacing.$

5577

Jan Ehrhardt @jehrhardt.dev · 13/09/2026

The stats are interesting. His own social media behavior even more. Nice that he prefers Bluesky to follow people.

000

Reposted by Jan Ehrhardt

Armin Ronacher @mitsuhiko.at · 12/09/2026

This is the type of post I might regret later, but I really had an urge to rant about the current discourse around pacing the frontier. lucumr.pocoo.org/2026/9/12/pd...

lucumr.pocoo.org

P(doom)

Are we all going to die?

6586

Reposted by Jan Ehrhardt

Dhruv Mehrotra @dmehro.bsky.social · 11/09/2026

NEW: Meta is facing a proposed class action alleging it used Facebook and Insta photos to build its NameTag face-recognition system and train AI models without consent. WIRED found code for NameTag hidden inside Meta’s AI app downloaded by tens of millions. Meta said the feature “doesn’t exist.”

wired.com

Meta Sued Over Training Data for Its AI and Face-Recognition Systems

The proposed class action alleges Meta illegally harvested people’s Facebook and Instagram photos to train its AI image-generation models and to build its unreleased “NameTag” face recognition feature...

16543237

Reposted by Jan Ehrhardt

n3rdopolis @n3rdopolis.bsky.social · 11/09/2026

Who would have thought this would actually be Omarchy? #Linux #OpenSource xkcd.com/528/

xkcd.com

Windows 7

0173

Jan Ehrhardt @jehrhardt.dev · 12/09/2026

Exactly this. Also @cachyos.org with niri as desktop 🚀

011

Reposted by Jan Ehrhardt

Zach Weinersmith @zachweinersmith.bsky.social · 11/09/2026

Very forceful statement by Terry Tao about AI and math. terrytao.wordpress.com/2026/09/11/a... Note, he is not claiming the models aren't real and powerful. He's talking about how that power should be used. This, I believe, is what all AI discussions should center.

terrytao.wordpress.com

A Severe Misalignment of AI in Mathematics

I am proud to be among the list of 25 initial signatories — all Fields Medallists — to the declaration below, which grew out of discussions between ourselves over the last week. We have…

19922230

Reposted by Jan Ehrhardt

👁️🍄🪸 @bloomfilters.plus · 11/09/2026

the mathematics award winning fields medalist and known himedanshi yu deng just said that if ai puts all mathematicians out of a job he's gonna go home and write yuri novels. girl please do it anyway i need to see what yuri a fields medalist mathematician can do omfg

212463614

Reposted by Jan Ehrhardt

Richard Jean So @richardjeanso.bsky.social · 11/09/2026

Cool new large grant opportunity from MacArthur Foundation on community-led AI, with one target area in the humanities + arts. Significant funding for projects that are trying to have real impact on human-centered and community centered AI development and use. humanityai.ai/open-call/?u...

humanityai.ai

Open Call for Proposals - Humanity AI

Open Call for Proposals: Communities Leading on AI $10M Investment in U.S.-Based Community-Led AI Initiatives Humanity AI, an initiative spearheaded […]

0146

Reposted by Jan Ehrhardt

Filippo Valsorda @filippo.abyssdomain.expert · 11/09/2026

Oh damn I had not seen the details of the MicroTik RCE: the client can send a public RSA key with correct N and e = 1 and the server will use it. Two primitives/protocol things that would have prevented it: if RSA was defined with a fixed e, and if SSH clients sent a key hash instead.

github.com

CVE-2026-67276 - GitHub Advisory Database

RouterOS does not compare the complete RSA public key...

26016

Reposted by Jan Ehrhardt

Mark Riedl @markriedl.bsky.social · 11/09/2026

Anthropic and OpenAI are warning about the existential threat of recursive self improvement. www.cnbc.com/2026/09/11/a... If you want a counterpoint, I've written on why the impacts of RSI might be spectacular in the short term but limited in the long term medium.com/@mark-riedl/...

cnbc.com

Why fears of AI self-improvement are causing ‘existential’ concerns at Anthropic and OpenAI

AI researchers are warning that faster AI self-improvement could eventually make advanced systems harder for humans to control.

1103

Reposted by Jan Ehrhardt

Brittany Ellich @brittanyellich.com · 11/09/2026

It seems like most of the "specialty" conferences about a specific language (JS, Ruby, Etc) don't really make sense anymore? Most of them are AI talks now

2132

Reposted by Jan Ehrhardt

Gergely Orosz @gergely.pragmaticengineer.com · 11/09/2026

A couple of trends I'm seeing across the industry. Which ones are you observing that I missed?

2120017

Reposted by Jan Ehrhardt

Saritha Rai @saritharai.bsky.social · 11/09/2026

DeepSeek newest V4.1-Flash is sending fresh jitters thru global rivals like Anthropic & OpenAI, and Chinese peers including Moonshot & Zhipu The AI model charges as little as a fraction of a cent per million tokens, ramping up pressure on rivals www.bloomberg.com/news/article... #AI

12111

Jan Ehrhardt @jehrhardt.dev · 11/09/2026

Well, time for rewrite by AI?

000

Reposted by Jan Ehrhardt

P(aul) Frazee @pfrazee.com · 10/09/2026

Just got called out on something and I think I should own up to it. We should've used erlang

3122911

Reposted by Jan Ehrhardt

Code BEAM @codebeam.bsky.social · 11/09/2026

@maennchen.dev & @peterullrich.com share how they found and fixed ~100 vulnerabilities across the top 1000 Hex packages, using AI to make the ecosystem safer at Code BEAM Europe 2026. 📅 21–22 Oct | Haarlem, NL | In-person & Virtual codebeameurope.com/#register

073

Reposted by Jan Ehrhardt

Andrew Lilley Brinker @alilleybrinker.com · 10/09/2026

#jj-vcs interoperability with Git continues to be key. The fact that ERSC can swap out the repo storage to not use Git, *and* can have a migration path to a future, better wire protocol than Git's, while preserving compatibility along the way, is amazing!

1385

Reposted by Jan Ehrhardt

Steve Klabnik @steveklabnik.com · 10/09/2026

We've been cooking at @ersc.io "What comes after git" ersc.io/blog/what-co...

ersc.io

What comes after git

Our thoughts about the future of version control

1625846