Windsurf is now Devin Desktop
Tracks the shift from a coding IDE into an agent command center, with local and cloud agents coordinated from one developer surface.
Read external articleAI articles archive
Review older AI development workflow articles without losing the timeline.
This archive continues the curated reading list for teams comparing AI coding tools, workflow controls, and model releases over time.
Updated Jun 17, 2026
Older coverage
Page 2
Older articles
Published before Jun 11, 2026
Use this page when you want earlier reference points for coding agents, model releases, and workflow control patterns without crowding the current-week page.
First head-to-head comparison of agentic AI applied to simulated Einstein Telescope data
Compares Claude Code and Codex on the same autonomous research workflow, which makes it useful for thinking about agent evaluation, verification, and reproducibility.
Read external articleWindsurf 2.0: Introducing the Agent Command Center and Devin in Windsurf
Good reference for agent orchestration patterns, especially when local IDE work and cloud agent delegation need to fit into one engineering workflow.
Read external articleDive into Claude Code: The Design Space of Today's and Future AI Agent Systems
An architecture-oriented read on Claude Code, permissions, context management, extensibility, and design tradeoffs in agent systems.
Read external articleApplying an Agentic Coding Tool for Improving Published Algorithm Implementations
Shows a two-stage research and implementation loop using Claude Code, with a strong emphasis on human validation, novelty, experimental correctness, and disclosure.
Read external articleAgentic Much? Adoption of Coding Agents on GitHub
A large-scale study of coding agent adoption across GitHub projects. Useful for teams that want evidence beyond vendor claims.
Read external articleDoes your code match your spec?
A Kiro article on property-based testing and executable specifications, directly relevant to review gates and evidence-led AI delivery.
Read external articleUsing Copilot Agent Mode to Automate Library Migration
A practical Copilot Agent Mode assessment showing why migration work still needs tests, acceptance criteria, and human review even when API usage migration succeeds.
Read external articleGitHub is launching a hub for multiple AI coding agents
Covers GitHub Agent HQ, third-party agents, Plan Mode, and code review controls, all relevant to teams comparing Copilot, Codex, Claude, and other agent workflows.
Read external articleThe Rise of AI Teammates in Software Engineering 3.0
Studies autonomous coding agents across real pull requests, including Codex, Devin, Copilot, Cursor, and Claude Code.
Read external articleIntroducing Claude 4
Model announcement covering Claude Opus 4 and Sonnet 4, coding performance, agent workflows, memory, parallel tool use, and Claude Code general availability.
Read external articleGitHub Copilot: Meet the new coding agent
GitHub’s introduction to assigning tasks to Copilot, reviewing pull requests, agent logs, MCP integrations, and human approval controls.
Read external articleOpenAI Codex
OpenAI’s coding agent introduction. Useful for understanding sandboxed code generation, testing, pull request preparation, and model-powered development tasks.
Read external article