On May 19, 2026, GitHub Copilot API requests experienced elevated error rates between 05:30 UTC and 14:50 UTC, with up to 13% of API requests failing and 24% of remote sessions failing to initialize. …
Based on a 2026 survey of developers, the article ranks AI coding assistants, with GitHub Copilot remaining the most widely adopted due to its improved multi-file awareness, while Cursor is highlighte…
GitHub has announced the general availability of remote control for GitHub Copilot CLI sessions, allowing developers to start work in VS Code or the CLI and continue it from their phone via the GitHub…
Arize AI today launched coding-harness-tracing, an open source tool that captures step-by-step traces of coding agent workflows across Claude Code, Cursor, Codex, GitHub Copilot, and Gemini CLI. The t…
Staff engineer Sean Goedecke now uses LLMs to produce entire pull requests in familiar code areas, relying on agents for every change and pushing PRs after a single editing pass as of early 2026. He a…
Zed now allows users to sign in with their ChatGPT account to access OpenAI models directly through its agent, with usage included in existing ChatGPT subscriptions rather than billed per token. The i…
A developer created a PowerShell script that integrates GitHub Copilot's CLI statusline with Oh My Posh, using a Catppuccin theme. The script parses JSON payloads from Copilot to display real-time tok…
Lamport Agent**, an AI-assisted tool that uses large language models (LLMs) to automatically generate formal specifications and detect race conditions from production codebases. It demonstrates the ag…
Recent breakthroughs in AI have enabled it to autonomously generate accurate formal specifications from large production codebases, as demonstrated when GitHub Copilot produced precise TLA+ specificat…
Experiment evaluating the AI coding assistant Windsurf on a complex, real-world task: implementing a group ID feature in Microsoft's RSL (Replicated State Library), a 7,500+ line codebase used in Azur…
Several key challenges faced by the Cursor development team, including improving context retrieval by training a custom reranker to filter 500k tokens down to 8k, creating a "copilot for edits" to han…