# Cursor

Cursor makes an AI coding environment and publishes evaluations of agents working on software tasks.

Type: organization. Context selected by Escape Meridian; inclusion is not endorsement.

[Collection](/entities/cursor)
[Original website](https://cursor.com/)

## Sources

- [Cursor / research and evaluations](https://cursor.com/evals): Follow model improvements into the rewards, environments and costs that shape useful software work.

## Perspectives


## Terminal notes

- [Learning from users also means choosing what counts as success](/notes/cursor-learning-from-use-2026.md): Cursor describes updating Composer from real interactions and reports better edit retention, fewer dissatisfied follow-ups and lower latency. It also describes incentives that rewarded broken tool calls and excessive clarification before the team changed them.

## Parallels

- [The answer is right. Was the question?](/parallels?parallel=right-answer-wrong-question): When producing an answer gets cheaper, where does the useful work move?

## Context references

- [https://cursor.com/](https://cursor.com/)
- [https://cursor.com/evals](https://cursor.com/evals)
- [https://cursor.com/blog/real-time-rl-for-composer](https://cursor.com/blog/real-time-rl-for-composer)

Latest published distillations: [Publications API](/api/pipeline/publications?entity=cursor)