Tau – Learn how coding agents are built
Tau is a Python coding agent built as a readable curriculum, exposing every layer of agent architecture with no hidden machinery.
Get hand-picked daily summaries of the best, most informative programming articles from around the web. Browse recent summaries, then jump to the original source when you want the full article.
Tau is a Python coding agent built as a readable curriculum, exposing every layer of agent architecture with no hidden machinery.
LLMs can draft impressive software fast, but production quality still demands iterative feedback loops, testing, and custom tooling.
CSS pseudo-classes increasingly replace JavaScript event listeners, and the upcoming event-trigger spec lets CSS animations respond to DOM events directly.
Vibe is a zsh plugin that converts plain English into ready-to-run shell commands via AI (OpenAI, Anthropic, or Ollama). Press Ctrl+G to generate an OS-specific command, then review and edit before running.
AI testing agents can't catch black swan bugs — catastrophic failures from untested system interactions that only experienced human engineers can anticipate.
JetBrains Air, the agent-first dev environment for running AI coding agents in parallel, now officially supports Windows.
This archive is built from SMRTR newsletter stories. Subscribe for hand-picked stories without the extra noise.
Build a Bloom filter from scratch in Python and understand false positives, sizing math, and real-world use cases like databases and safe browsing.
Outer Shell turns servers into graphical app environments accessible over SSH, using Unix sockets and simple HTTP servers with no encryption overhead.
Agentic Orchestrator automates the full feature-to-PR pipeline with parallel AI agents, isolated worktrees, and optional human approval gates.
DocumentDB is a MongoDB-compatible, MIT-licensed document database built on PostgreSQL with native BSON, vector search, and a Kubernetes operator in preview.
Stop waiting for smarter AI to fix your bottlenecks — build ergonomic, human-in-the-loop workflows with the tools you have today.
Cursor Mobile lets developers manage AI coding agents from their phones, reflecting a shift away from desktop setups toward mobile-first agent oversight.
Pure is a terminal-based word processor for Markdown and HTML that edits semantic document structure instead of raw markup syntax.
Dribble is an open-source SQL IDE with a built-in AI analyst that writes and runs queries against your Postgres database.
No official Ruby SDK for Gemini? This guide shows how to build a production-grade Rails agent using the Interactions API with Faraday.
Agent Rigor is an open-source framework that forces AI coding assistants through a strict six-phase workflow—planning, testing, and verifying fixes—to prevent the "Doom Loop" of confidently repeated failures.
One slow dependency can freeze your entire API — here's how timeouts, retries, and circuit breakers prevent cascading failures in production.
Prompt injection lets attackers hijack LLM-powered apps using plain English — here's how it works and how to defend against it.
Microsoft's MAI-Code-1-Flash is now live for GitHub Copilot Business and Enterprise, offering fast, low-latency AI code generation for large dev teams.
AgentBrush gives AI coding agents persistent visual taste, keeping every generated image on-brand across Claude Code, Cursor, Windsurf, and more.
A clear, math-light breakdown of how transformers work — from word embeddings to attention mechanisms — using one running example throughout.
A free, open-source 30-topic Laravel performance guide covering algorithmic pitfalls, production disasters, and architectural patterns for high-scale systems.
Murmur lets Claude, Codex, Gemini, Cursor, and Copilot share a local chat room via MCP so they can delegate tasks to each other directly.
asciigraph is a Go package that renders ASCII line graphs in the terminal with multi-series, color, gradients, and real-time CLI streaming support.