#agentic coding
3 articles tagged agentic coding
Tagged articles
Articles tagged agentic coding, newest first.
AI in PracticeOrca Agent IDE: Why a Terminal Alone Falls Short for Claude Code Workflows
MIT-licensed agent IDE that runs Claude Code, Codex, Gemini and other CLI agents. CLI agent backends run on flat-rate monthly plans instead of per-token API billing. One agent per Git worktree, created in a click from GitHub Issues, Jira or Linear. On a remote VPS, Chat Mode types locally and sends whole prompts, hiding SSH lag. Built-in Chromium browser lets you annotate a live page and send it as one prompt.
AI in PracticeLauren Tan's Six AI Agent Habits: Fix the Setup, Not the Chat
Lauren Tan reports merging 2,000 to 2,500 pull requests a month with coding agents. Start with a tiny bot team and zero skills, then codify only repeated failures. Make the agent restate your goal in its own words before it writes code. Fix errors with architecture, lint rules and tests before writing prompt rules. Her output rests on roughly 15 billion tokens a day, so avoid 15-minute routines.
AI in PracticeAI Coding Agents at Uber, Ford and Databricks: More Code, Same Bottlenecks
AI agents multiply code output but review and release gates now cap delivery speed. Gains concentrate in internal tools and framework upgrades, not consumer features. Roughly 70 percent of human code reviews are rubber stamps, by one estimate. Token ceilings, daily tripwires and manager approvals now govern AI spend. Use scripts for chores and carve 5 percent of engineering time for prompt tuning.
Get new articles by RSS or podcast
Add the RSS URL to your RSS reader or the podcast URL to your podcast app to get every new article as soon as it is published. The podcast reads out the headline and key points.
How to subscribe