Profile
Its so annoying how Claude always adds so much unneeded comments in Code but also this too long documentations and READMEs. I hate how bad LLMs are at removing code or keeping their own mess up to date. Nice to see a study about this topic: Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding
Building Software Is Learning - true 100%
Best comparison of the 2008 financial cricis and the AI boom. Super interesting read.
Cool in-depth blog post about eliminating branches C++ in loops. Written by one of the authors of Ada, a high-performant compliant URL parser used at Internet archive, Node.js, Cloudflare and more.
How lina used an expired nameserver to take over the e164.arpa zones for multiple territories and logged hundreds of thousands of phone calls to military bases. I never heard of e164.arpa before, so its cool to read about it here.
Code Isn't Free — Mario Zechner on the Hard Truths of Coding With AI (creator of Pi) Have not fully heard it, but always interested in peoples opinion who build AI tools but are not trying to sell you the AI hype like the all these ex. crypto bros, who are now AI bros.
How to Write a Paper in a Weekend (By Prof. Pete Carr) It‘s a ten year old video, but still super interesting and teaches you an algorithm how to effectively write. „Writing is the most exacting part what we do as a scientist.“
The "Big Short" Guy Who Called (And Cashed In On) The 2008 Housing Bubble Just Bet Against the AI Bubble While Betting on Financial Armageddon In The US Michael Burry looked at the most hyped trade in human history and said “nah.” You should probably pay attention.
Building pi in a World of Slop: "All I wanted was a shitty coding agent that is truly mine. And I’d have loved to just tell you why and how I built pi. But then Peter decided to make it the agentic core of OpenClaw. And now pi is collateral. So yes, this is a talk about pi. But it is also a talk about how agents are destroying OSS, how I deal with that, and a plea to slow the fuck down." by Mario Zechner
ShieldFront is an open-source creative technology intervention developed to protect human writing from unauthorized AI training. Check it out its kinda interesting how they do it.
Short introduction to SIMD (Single instruction, multiple data) by Mitchell Hashimoto
Detailed report by Hugging Face after the OpenAI hack: Anatomy of a Frontier Lab Agent Intrusion: A Technical Timeline of the July 2026 Incident
Follow Up video to "I Tried To Make Something In America" by Smarter Every Day a year after the first video.
I Tried To Make Something In America (The Smarter Scrubber Experiment) by Smarter Every Day. Interesting video to help understand the supply chain needed to create even a simple product as a scrubber for a grill.
Introduction to Zig, a project-based book by Pedro Duarte Faria
A Sovereign, Open-Source Foundation Model for German and English
/e/OS is a complete, fully “deGoogled”, mobile ecosystem
Mini-Coding Agent - Minimal and readable coding agent harness implementation in Python to explain the core components of coding agents.
to be read: Apps After Agents by Kyle Mathews
to be read: Harness Engineering for Self-Improvement by Lilian Weng
If you want to see how Google AI search changes the internet, I recommend the blog article by Amanda: How Google and AI are killing travel blogs like mine
The Man Who Killed Google Search by Ed Zitron
Looks super interesting: Urban Analytics with Python — An Introduction: Spatial Data, OpenStreetMap, and the Basics of Urban Analytics
GLM and DeepSeek Are Catching Frontier Models at Finding Vulnerabilities
a11y.quest a free accessibility quiz to help you check your accessibility knowledge, by Dave Davies. You get 128 hand written questions across WCAG 2.2, ARIA, semantic HTML and contrast. And for each, an explanation for each answer.
What I learned building an opinionated and minimal coding agent by Mario Zechner
Stéphanie Walter's UX Research and Design Blog - Practical expert insights and curated content (resources, tools, etc.) on UX research, inclusive design, enterprise UX, accessibility, and more, to help you growth as a designer.
AI-Generated UI Is Inaccessible by Default - A five-layer enforcement system for semantic correctness in LLM-generated React components
Here’s how to instruct a LLM to reference the ARIA Authoring Practices Guide by Eric Bailey
Using Local Coding Agents - Using Open-Weight Models in Local Coding Harnesses as an Alternative to Claude Code and Codex Subscriptions by Sebastian Raschka
Build a Reasoning Model (From Scratch) - Book - by Sebastian Raschka (also need to still read it)
Rust to WebAssembly the hard way
Prompting 101 | Code mit Claude by Hannah Moran and Christian Ryanop from Anthropic - have not watched it yet, but it was suggested to me - let's see if it is interesting (already 10 months old, and we know things change fast currently)
the true reason C++ always wins by LaurieWired
Nice blog by Dinh Doan Van Bien - has a series about building Supabase froms cratch or his experiements and views on vibe coding or how he build a disk space analyzer for macOS
I Could've Rickrolled the Entire FIFA World Cup. All I Needed Was My ID by BobDaHacker
I think I want to buy this: Build a Large Language Model (From Scratch) I <3 things being explained from scratch.
A Visual Guide to Attention Variants in Modern LLMs - seems interesting (not ready yet)
Building an Agent from Scratch (except the LLM) - one of the most interesting articles I read. It actually explains how to implement a fully functional code-editing agent with less than 400 lines of Go.
Hacking Google with A.I. for $500,000 - What happens when you unleash an AI across all of Google's infrastructure? 1,500 APIs, 3,600 keys, and $500,000 in bounties later.
A small quiz to train your judgement which UI animation feels better. And it gives a great explanation for each why it feels better or worse.
How LLMs Actually Work - nice blog article about LLMs and the transformer-family architecture on a higher-level