It's a pitty that the author did not test Cursor with Claude 3.7 Sonnet. That way he would have separated the improvements due to the model used from the improvements due to the better DX, system prompts, RAG, Agentic workflows, etc.

- No significant improvement in code generation over Cursor.

- Cursor outperformed Claude on pure backend changes.

- Claude excelled in handling integrated backend and frontend changes.

- Both tools performed well on pure frontend changes.