Agentic Coding: How Google’s Jules Compares to Claude Code

After my successful use of Claude Code, I was keen to try other equivalent agentic large language model (LLM) tools that aren’t IDE-based. These have better ways to interact with a plan than staring at files. Because a plan can be expressed as an ordered set of well-known coding practices, an LLM just acts as a code wrangler — which we know it does well.

The Google project Jules describes itself as “an async development agent — Jules tackles bugs, small feature requests, and other software engineering tasks, with direct export to…