A CPU that runs entirely on GPU
A CPU that runs entirely on GPU — registers, memory, flags, and program counter are all tensors.Every ALU operation is a trained neural network.Addition uses Kogge-Stone carry-lookahead. Multiplication uses a learned byte-pair lookup table.Bitwise ops use neural truth tables. Shifts use attention-based bit routing. No hardcoded arithmetic. Quick Start pip install -e ”.[dev]” # Run […]
Nobody Gets Promoted for Simplicity
“Simplicity is a great virtue, but it requires hard work to achieve and education to appreciate. And to make matters worse, complexity sells better.” — Edsger Dijkstra I think there’s something quietly screwing up a lot of engineering teams. In interviews, in promotion packets, in design reviews: the engineer who overbuilds gets a compelling narrative, […]
California’s Digital Age Assurance Act, and FOSS
These solely represent my own interpretation and opinions. Some parts might be wrong — actually — some parts will be wrong. Please tell me somewhere. I am not a lawyer. This is provided solely for the purposes of general information and does not constitute legal advice, guidance, or counsel. No attorney-client relationship is established by […]
Speculative Speculative Decoding (SSD)
Abstract:Autoregressive decoding is bottlenecked by its sequential nature. Speculative decoding has become a standard way to accelerate inference by using a fast draft model to predict upcoming tokens from a slower target model, and then verifying them in parallel with a single target model forward pass. However, speculative decoding itself relies on a sequential dependence […]
Just Send the Prompt
Are you about to copy and paste the output of an LLM into an email, comment, ticket, or anything that another human is expected to read? Don’t! Just send the prompt There’s no point to what you’re doing. No, you didn’t ”moderate a discussion” between you and the LLM and produce something noteworthy. No, your […]
Mac external displays for designers and developers, part 2
Since writing about Mac external displays in 2016, not much has changed. LG, Dell, Samsung, and other display makers have either never catered for the specs many Mac designers and developers want, or they’ve reluctantly produced products that have been short-lived or compromised. In 2019, Apple announced their return to the display market, with an […]
Graphics Programming Resources
Lecture 19: Variance Reduction (CMU 15-462/662) An overview various variance reduction techniques for Monte Carlo rendering, covering bidirectional path tracing, Metropolis-Hastings algorithm, MIS, stratified Sampling, low-discrepancy sampling, blue noise, photon mapping, and finite element radiosity
Weave – A language aware merge algorithm based on entities
Resolves merge conflicts that Git can’t by understanding code structure via tree-sitter. The Problem Git merges by comparing lines. When two branches both add code to the same file — even to completely different functions — Git sees overlapping line ranges and declares a conflict: theirs — function `process` (both modified)” readability=”34″> > theirs — […]
TikTok will not introduce end-to-end encryption, saying it makes users less safe
”We know just how risky end-to-end-encrypted platforms can be for children, preventing the detection of child sexual abuse and exploitation and contributing to a worrying global decline in reports,” said Rani Govender, its associate head of policy for child online safety.
Marcus AI Claims Dataset
Gary Marcus is the most prolific AI skeptic on the internet. Since May 2022, he’s published 474 posts on Substack making claims about AI’s limitations, the companies building it, and where the industry is headed. We extracted every testable claim. 2,218 of them. Scored each one against the evidence as of March 2, 2026. Here’s […]