The latest blog posts to take off on Hacker News. 3,090 posts.
I compiled a RISC-V CPU into Qwen 3.5 4B's weights, using its Gated DeltaNet layers as RAM. It runs C and MicroPython inside the forward passes, and Qwen uses it for exact arithmetic.
AI agents write code faster than anyone can read it. What engineers still need: fundamentals, breadth, and the willingness to dive in when it breaks.
Agents made implementation cheap, so I started doing far more of it in parallel. My GitHub history since 2019 shows what that did: as VP of Technology I am back at my full-stack engineer pace, with bursts, burnout and a low season that is no longer low.
An update to Brooks's Law in the age of AI.
Over 11 hours of coding videos on evaluation, inference-time scaling, reinforcement learning, and distillation, with Python and PyTorch.
What is “the true cost of a token”?
As models improve, they bring surprising challenges for those building apps and harnesses. How context engineering has changed this year, and where it’s headed.
America is Losing Hundreds of Thousands of Jobs in Media, Film & the Arts. How Much is AI to Blame?
A 360M model that pretends to be a terminal, and remembers what you did
LLM-based agents can do a lot of tasks. Or, at least, they can appear to do them. For example, they write code, summarize papers, plan trips, and fill in forms. If you ask one for a probability or a confidence score, it will give you one of those too. I have seen many agent designers use LLMs to generate confidence…
Welcome to the digital home of Daniel Ball. Explore my portfolio to see the projects I’ve worked on and learn more about my professional journey.
These past few weeks, I’ve been conducting interviews, and I often open with “make sure you articulate your thoughts out loud, so I can follow along with your thinking”. This is primarily because I…
Agents are structurally well suited to monitoring production systems.
Of all the forms of tech debt, decoherence is the worst. It compounds exponentially with a system's rate of change, and AI is now producing it faster than ever, leading us deeper and deeper into the abyss...
A working tmux setup for running Claude Code and other AI agents: session-per-project naming, pane layouts, send-keys for scripting an agent, capture-pane for reading one back, and the status-line monitor. Plus the limits you hit at four agents.
I think the title captures the idea fully but I'll dig into it because this ain't Twitter. You're working an idea, a piece of code, or a multibillion a year SaaS app which is probably saving the world. Maybe you're automating the cloning process of grabbing
Every layer is a memory store
Answering questions and concerns.
The enormous advance in AI capabilities over the last several years has in large part been driven by the availability of vast amounts of training data.
A cautionary tale.
It is important to provide credit.
A little AI tells detector helps us promote good writing.
Why do LLM pelicans look so alike? A visual look at diversity collapse and our research on creativity in UI design.
AI is now able to decompile games and software in a matter of days. That's scary... but kind of exciting.
Exploring practical ways to make your game more legible to AI agents and how to make agentic iteration faster.
The OpenAI algorithm for the FFT is less a breakthrough and more a malicious exploit.
I turned my unused Kindle into a photo frame with an AI agent.
A response to Rich Hickey's "'AI' is Not a Tool": automation and tool use can coexist, and wielding an LLM well requires skill, feedback, and judgment.
I recently read End Times Fascism and the Fight for the Living World by Naomi Klein and Astra Taylor. It's an excellent book and I highly recommend that everyone read it. Tech oligarchs make up an important component of the end-times-fascist coalition. One thing that stood out
Kelly predicted AI heavyweights while advocating for a future of open weights
