mattwood.fyi
mattwood.fyi
Links indicate relevance, not agreement. How to use this site →
Techniques for integrating computer and browser use capabilities into a cloud software factory, enabling AI agents to reproduce bugs, verify fixes, and confirm new features match specifications. The approach covers how computer use adds value across triage, implementation, and code review phases of an agentic development workflow.
An essay arguing that as AI collapses the cost and effort of building software, the filtering function that effort once provided disappears, leaving "taste" — the judgment of what deserves to exist — as the remaining meaningful differentiator in software craft.
AMD's acquisition of AI chip startup Taalas aims to enhance inference performance by physically encoding AI models directly into silicon hardware. The deal represents AMD's effort to compete in the AI accelerator market by leveraging Taalas's approach of optimizing chips at the hardware level for specific AI workloads.
OpenAI's work on improving the GPT-4.5 or a related model's performance on solving problems, likely focusing on enhancements to reasoning, accuracy, or problem-solving capabilities within ChatGPT. The content likely details technical improvements, benchmark results, or methodology changes made to advance the model's abilities.
An exploration of how the economic concept of Pareto efficiency can be applied to optimizing character and kart builds in Mario Kart 8, using the game's multiple competing statistics (speed, acceleration, handling, etc.) to identify dominant choices and eliminate suboptimal ones. The piece uses interactive visualizations to demonstrate how the Pareto front helps narrow down thousands of possible build combinations to a set of objectively non-dominated options.
Important work.
Discovery Loop is an AI company focused on automating scientific and engineering experimental loops, using frontier AI models and large-scale computational infrastructure to parallelize and accelerate the process of discovery. The company, founded by Jeff Dean, Sanjay Ghemawat, Quoc Le, and Oriol Vinyals, begins with automating machine learning research before expanding to tackle broader scientific grand challenges.
Construction of a production-grade agentic harness for LLMs, covering typed tool validation, parallel execution via dependency graphs, multi-tier memory, verification hierarchies, role separation (Planner/Worker/Critic), and budget controls. A city comparison agent serves as the running example to illustrate how these primitives compose into a reliable, debuggable system.
TL;DR: Scientific invention requires manipulative abduction and physical simulation
An argument that intelligence and technological capability are often not the primary barriers to real-world progress, using examples like medical regulation and housing policy to illustrate how political will, institutional constraints, and regulatory bottlenecks frequently matter more than raw intelligence or AI advancement.
A technical article published in ACM Queue likely covering a specific topic in computer science, software engineering, or systems design, as is typical of the publication's focus on practical and research-oriented computing topics.
The myths are:
A discussion of Pi, a minimalist AI coding harness with only 4 tools and under 1,000 tokens in its system prompt, which achieves industry-leading performance at lower cost by keeping context lean and avoiding excessive orchestration layers. Case studies from Databricks and Shopify illustrate how Pi's minimal design outperforms more complex coding agents on real-world tasks.
Pi is the coding harness that chooses minimalism on purpose. It comes out of the box with only 4 tools, and its system prompt and tool definitions come in below 1,000 tokens. The idea being that most work can be done with the basics, and if you want more, build it.
An analysis of how previous technology revolutions (PCs, the internet, smartphones) typically increased workloads and benefited companies more than employees, and why AI may be different by actually performing work rather than just accelerating it. The piece argues that while AI is currently causing workplace disruption, it has the potential to free up time for creative thinking and collaboration rather than simply raising productivity expectations.
An exploration of how humans solve Coverage Path Planning problems, using an interactive lawn-mowing experiment where tens of thousands of participants found near-optimal paths, compared to the mathematical challenge these routing problems pose for computers. The piece examines why humans are surprisingly efficient at spatial navigation tasks like mowing or vacuuming despite the astronomical number of possible route combinations.
From the article, a good description of algorithms and heuristics.
A slight detour, if we may. There are two main ways to tackle a problem like this. Simplifying things a bit, there are algorithms which guarantee the optimal path, and heuristics which use shortcuts to find a “good enough” path, fast.
Also, this domain is a keeper.
Shieldstral is a safety-focused AI model or tool introduced by Mistral AI, designed to provide content moderation and security capabilities for AI applications. It is part of Mistral's expanding lineup of specialized models and tools for enterprise and developer use.
The Mistral Cinematic Model Universe continues to expand.
Kiro Crew is an open-source, persistent AI development workspace that retains memory and context across sessions, learns from user workflows, and coordinates autonomous agents to handle tasks like issue triage, CI/CD monitoring, and scheduled jobs — even when the user is away. It includes multi-layered security, a knowledge graph with vector search, and editable lessons and skills stored as Markdown files.
8 · kiro.dev →Steve Yegge discusses AI coding agent techniques, particularly using "loops and graphs" harnesses to tackle large problems autonomously, and introduces his custom harness called Wheelhouse built for his long-running MMORPG project Wyvern. He argues that reusable harness frameworks are a dead end and that effective AI harnesses must be bespoke and tightly integrated into the specific application being built.
Don't be special, stay out in front, and you will see the future clear as day.
Good advice.
15 · yegge.ai →TerminalWidget is a macOS, iOS, and iPadOS app that allows users to display terminal command output, scripts, API data, and Shortcuts directly in native widgets, with support for rich text formatting, progress bars, charts, sparklines, and images. It includes a full CLI, AppleScript support, URL scheme automation, and syncs across Apple devices via iCloud.
Great way to enable your agent to share status, notifications, data, etc. Cool.
"Everything I Know" refers to a series of recordings by Buckminster Fuller, a comprehensive collection of his ideas and philosophy captured over an extended lecture session. The Buckminster Fuller Institute provides information about this landmark series in which Fuller synthesized his lifetime of thinking on design, systems, and humanity's future.
So great.
Marin is an open collaborative lab for building foundation models from scratch, sharing all code, data, experiments, and results transparently in real-time. It invites open-source contributors to participate in model training, architecture research, and experiments, with publicly documented models like Marin-8B and Marin-32B that compete with leading open-weight models.
Open Athena is a nonprofit organization that partners with academic institutions to develop open-source AI foundation models by providing engineering talent, compute resources, and coordination. Current projects include large language models with Stanford, plant genomics DNA models with Cornell, and protein generation models with MIT.
Bonsai is an OCaml library by Jane Street for building dynamic web applications using Js_of_ocaml. It provides a framework for creating interactive front-end UIs compiled from OCaml to JavaScript.
Domain expertise is the most important factor in effectively using LLMs, as demonstrated by Terence Tao's mathematically sophisticated ChatGPT conversation — skilled users can steer models more precisely, recognize flawed outputs, and suggest better approaches because they understand the subject matter deeply. Unlike generic prompting tips, this advantage cannot be replicated without genuine knowledge of the domain.
OpenAI's work on ten notable advances or breakthroughs in mathematics, likely highlighting contributions made by AI systems such as their models in solving or progressing on significant mathematical problems and conjectures.