Full-text search across 30593 articles. Combine with topic and date filters; results sorted by relevance.
An agentic loop is an iterative execution cycle where an AI coding agent autonomously plans, executes, and evaluates actions — such as writing code, running tests, and fixing bugs — without waiting for human input. This …
BrassCoders finds that autonomous coding agents, such as Claude Code and SWE-agents, bypass human code review, allowing security vulnerabilities like hardcoded credentials to go undetected. The company's research shows A…
AI-assisted development tools are splitting into two distinct categories: coding agents and product agents. Coding agents like Cursor and GitHub Copilot help developers edit code within existing projects, while product a…
GitGuardian has introduced an MCP server that integrates directly into AI coding agents like GitHub Copilot to enforce real-time security checks on generated code. The tool addresses the challenge of securing AI-generate…
Thomas Depierre argues that reviewing AI-generated code is not a viable defense for using LLM coding assistants, citing scientific evidence that they do not help developers write code better or faster. He contends that t…
A developer argues that the Figma-to-code pipeline is overhyped, with naive AI agents producing code that invents components, hard-codes colors, and mixes conventions. However, monday.com's design-to-code pipeline demons…
Anthropic's Claude Code and OpenAI's Codex represent two distinct approaches to AI-assisted coding, with Claude Code designed for real-time collaborative "steering" and Codex built for independent task "dispatching." Cla…
Eight months ago, software engineer Malte Skarupke wrote roughly 90% human code and 10% AI code; now his code is about 90% AI-generated. In a recent optimization of a matrix-multiply kernel using Nvidia's Cluster Launch …
Software engineer JP Hutchins is developing an open-source agentic code review tool that interacts with code changes rather than just reading them, and he questions whether the model can survive a post-token-subsidy worl…
Andrej Karpathy, who coined the term 'vibe coding,' said on the No Priors podcast in March 2026 that he has not typed a line of code since December 2025, reflecting a dramatic shift in software engineering driven by AI c…
The article explains that the primary goal of code review is to help the reviewee improve their code, requiring a constructive approach that encourages engagement. It notes that in 2026, LLMs have made code review easier…
Cursor, a fork of VS Code, and Claude Code, a CLI agent from Anthropic, differ fundamentally in architecture and use case, with Cursor offering faster inline edits for feature building and Claude Code excelling at autono…
Developers using LLMs to write code risk losing familiarity with their own codebases, warns a software engineer in a personal essay. As LLMs generate more code, engineers struggle to remember decisions, locate code, and …
Deslop, a live duplicate-code analysis server for C# with a VS Code extension, LSP, MCP server, and CLI, detects exact, renamed, and near-miss duplicates using syntax-aware analysis and optional local embeddings. The too…
A 2026 analysis of AI-assisted code development reveals that while AI agents dramatically increase code output, human code review has become the primary bottleneck, with review duration surging 441.5% and defect rates ri…
OpenAI's Codex CLI and Anthropic's Claude Code are terminal-based AI coding agents built on different models, with Claude Code excelling at deep reasoning for complex refactoring and ambiguous requirements while Codex CL…
On February 2, 2025, Andrej Karpathy coined the term "vibe coding," a development style where users describe software in plain English and AI writes the code, with Collins Dictionary naming it the 2025 Word of the Year. …
AI coding tools like Claude Code, Cursor, and Codex are generating functional code that often contains security vulnerabilities, according to multiple reports from Veracode, CodeRabbit, and Apiiro. The Veracode 2025 GenA…
AI-assisted pull requests contain 1.7 times more issues than purely human-written code, according to CodeRabbit's State of AI vs Human Code Generation report analyzing 470 real-world GitHub pull requests. The report foun…
Developers who rely on AI coding assistants like GitHub Copilot must write clean code because sloppy code trains the LLM to generate even sloppier code, creating a feedback loop that turns repositories into technical deb…