if an LLM wrote a significant amount of the implementation, apparently the human directing it no longer counts.
I Still Built It — Ewan's Corner
if an LLM wrote a significant amount of the implementation, apparently the human directing it no longer counts.
Lisp Against the (LL)Machine
Every Lisp, Scheme, and Lisp-adjacent project listed is a non-toy implementation that is at least somewhat active. When I first wrote this article (2026-03-12), most Lisps did not have a policy or even public stance on LLM contributions. In such cases I posted to their bug trackers or discussion lists to ask. Most maintainers kindly responded.
David Hendrickson on Twitter / X
🌞This is big Local AI news! A new open-source Computer-Use LLM has just launched. Holo 3.1 is H Company’s (🇫🇷) new local computer-use agent model that beats Qwen3.5-397B, Kimi-K2.5, and Sonnet 4.6!Since it is built for local deployment → ⬩ Runs fully on your machine… https://t.co/CpOEsuWN2k pic.twitter.com/w39iOh7cO1— David Hendrickson (@TeksEdge) June 2, 2026
Dan Shipper 📧 on Twitter / X
this is true and is a big reason why you don’t need to be a highly technical researcher to use LLMs in surprising and novel ways https://t.co/TuxNzXzToU— Dan Shipper 📧 (@danshipper) July 27, 2025
Karpathy's LLM Wiki: The Complete Guide to His Idea File
Karpathy's follow-up gist went viral. Complete breakdown with implementation examples.
The Bitter Lesson of LLM Extensions
From ChatGPT Plugins to Agent Skills, a look at how we've been trying (and failing) to extend LLMs for the last three years.
"Useful" is not sufficient
So Linus Torvalds, head of the Linux kernel development, put his foot down on the Linux Kernel development mailing list when someone was bringing up criticism of LLMs: “Linux is not one of those anti-AI projects, and if somebody has issueswith that, they can do the open-source thing and fork it. Or just walk away. […]
GitHub - humanlayer/12-factor-agents at sidebar
What are the principles we can use to build LLM-powered software that is actually good enough to put in the hands of production customers? - GitHub - humanlayer/12-factor-agents at sidebar
Paul Iusztin on Twitter / X
Using LLM wikis for agentic coding is pure gold. It's so powerful it feels like cheating. Here is how I used it to develop a coding harness from scratch:1. I ingested multiple coding harnesses repositories into the LLM wiki (e.g., opencode, pi, hermes, etc.) pic.twitter.com/7WE7NI9tnN— Paul Iusztin (@pauliusztin_) June 29, 2026
crawshaw - 2026-02-08
I wrote up my experiences programming with LLMs a bit over a year ago, and updated it for the world of agents eight months ago. A lot has changed since then, so here is an update.
I rebuilt Karpathy's LLM Wiki gist: what's missing
Karpathy's original LLM Wiki gist, explained and rebuilt as a working system. 44 open-source commands, plus the five things the gist leaves out.
Why your local LLM feels dumber than it is
Quick Introduction We have all been on forums, chats, reddit, discord, youtube, or somewhere and heard “Oh! Model XYZ is AMAZEBALLZ!zomgwtfbbq” then downloaded it (or more likely, some quantized form of it) and said “eww… This sucks!” This post is going to be a rather technical series of experiments to demonstrate the impact of implementation-specific hazards with inference. I will be using the term “reference implementation” to describe the lab that published and offers first-party hosting of ...
I accidentally turned LLM memory into program analysis :: pwning.systems
Why I stopped trying to give LLM agents a better memory and instead built Lemmalog, a Datalog engine that maintains an agent's knowledge as analysis state, with provenance, retractions and incremental evaluation, plus what happened when I benchmarked it on LongMemEval and LoCoMo.
Andrej Karpathy on Twitter / X
LLM Knowledge BasesSomething I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating…— Andrej Karpathy (@karpathy) April 2, 2026
Also with the obligatory basic bugs common to LLMs "The author of this library would like everyone to know that he was, in fact, well-aware of this requirement, thought about it a lot while designing the library, and then, somehow, forgot" nvd.nist.gov/vuln/detail/cve-2025-4143 nvd.nist.gov/vuln/detail/CVE-2025-4144
NVD - cve-2025-4143
nvd.nist.govI haven't, but I don't know that it's that kind of model. It's really different than an LLM, since if you want an answer from it you can only ask it multiple choice, true or false ( with confidence ), or "tell me the score of X" questions. So there's no generating proof code or anything directly.
