Machines

An eight-month study found that these tools made productivity surge—as well as cognitive fatigue, unsustainable hours, and other problems.

1 skill, 17 commands, and curated anti-patterns for impeccable frontend design. Works with Cursor, Claude Code, Gemini CLI, and Codex CLI.

A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes...

Self-improving AI systems aim to reduce reliance on human engineering by learning to improve their own learning and problem-solving processes. Existing approaches to self-improvement rely on fixed, handcrafted meta-level mechanisms, fundamentally limiting how fast such systems can improve. The Darwin Gödel Machine (DGM) demonstrates open-ended self-improvement in coding by repeatedly generating and evaluating self-modified variants. Because both evaluation and self-modification are coding tasks, gains in coding ability can translate into gains in self-improvement ability. However, this alignment does not generally hold beyond coding domains. We introduce \textbf{hyperagents}, self-referential agents that integrate a task agent (which solves the target task) and a meta agent (which modifies itself and the task agent) into a single editable program. Crucially, the meta-level modification procedure is itself editable, enabling metacognitive self-modification, improving not only the task-solving behavior, but also the mechanism that generates future improvements. We instantiate this framework by extending DGM to create DGM-Hyperagents (DGM-H), eliminating the assumption of domain-specific alignment between task performance and self-modification skill to potentially support self-accelerating progress on any computable task. Across diverse domains, the DGM-H improves performance over time and outperforms baselines without self-improvement or open-ended exploration, as well as prior self-improving systems. Furthermore, the DGM-H improves the process by which it generates new agents (e.g., persistent memory, performance tracking), and these meta-level improvements transfer across domains and accumulate across runs. DGM-Hyperagents offer a glimpse of open-ended AI systems that do not merely search for better solutions, but continually improve their search for how to improve.

Uber’s uSpec uses AI agents and the Figma Console MCP to automate design specs, cutting documentation time from weeks to minutes. Integrated with the Michelangelo platform, it uses a GenAI Gateway for

When faced with an unexpected tool response, without any adversarial attack, Gemini 3 deliberately and covertly violates an explicit system prompt rule.

In this codelab you’ll learn on how to start working with Cloud Services using Google Cloud Google Cloud MCP services. It walk you through necessary setup and configuration steps and then help to migrate a local application to Google Cloud using Gemini CLI and Google Cloud MCP

We've moved past prompt hacks. It's time to invest in durable context systems.

‘If the maintainers of small projects give up, who will produce the next Linux?’

State-of-the-art TTS model under 25MB 😻

Autonomous experiment loop extension for pi

PR from Shopify CEO Tobias Lütke against Liquid, Shopify's open source Ruby template engine that was somewhat inspired by Django when Tobi first created it back in 2006. Tobi found …

Spotify credits Claude Code and its internal AI system Honk with speeding up development.

The open protocol for machine-to-machine payments

We’re launching the Machine Payments Protocol (MPP), an open standard, internet-native way for agents to pay—co-authored by Tempo and Stripe. Businesses on Stripe can accept payments over MPP in a few lines of code using our PaymentIntents API.

In 2005, a freestyle chess tournament was held online. Freestyle means anything goes - players can use chess engines, databases, teammates, whatever they like.

Explore the insights from our biggest research ever.

Tool-Hopping Is Why You Feel Crazy. Professionals Figured This Out Decades Ago.

ML, Biotech, Hardware, and Coordination Problems. Sometimes I write about hard problems and how to solve them.

Software engineering is not about writing code anymore. It is about building the factory that builds your software.

Learn how tracer bullets can help you control AI code quality by building small, end-to-end features and getting feedback early instead of bloated solutions.

Exploring the chardet v7.0.0 controversy: Can an AI rewrite legally 'launder' a library from LGPL to MIT?

System prompts matter far more than most assume. A given model sets the theoretical ceiling of an agent’s performance, but the system prompt determines whether this peak is reached.

“” is published by Steve Yegge.

The latest installment from Steve Yegge on viiiiibe coding and what that means for developer jobs.

This is a sample AGENTS.md file from my classes. GitHub Gist: instantly share code, notes, and snippets.

In this essay, Carson Gross discusses his advice to young people interested in computer science worried about the future given the advancements in AI.

The technology has resulted in a host of cutting-edge AI applications — but its real power lies beyond text generation

I’ve started a new project to collect and document Agentic Engineering Patterns—coding practices and patterns to help get the best results out of this new era of coding agent development …