
No need for software engineers? Cursor launches new tool that lets AI coding agents run on their own
India Today
Cursor has introduced a new feature called Automations that allows AI coding agents to start tasks automatically inside a developer's workflow. The system aims to reduce the need for engineers to constantly prompt and monitor multiple AI tools.
The growing use of AI in software development has started to change what a typical workday looks like for programmers. Engineers are no longer just writing code line by line. In many teams, they now work alongside AI agents that can generate, review, and analyse code on their own. But as more of these tools enter the workflow, developers are also facing a new problem: managing them. A single engineer may end up supervising several AI agents at the same time, making attention and coordination a key challenge.
Cursor, the company behind an AI-powered coding environment, believes automation can help solve that. The company has introduced a new feature called Automations, designed to allow AI coding agents to run tasks automatically inside a development workflow instead of waiting for engineers to manually trigger them. Cursor has launched a new feature called Automations that allows AI coding agents to run tasks automatically inside a developer’s workflow, reducing the need for engineers to manually prompt and monitor multiple AI tools.
In many AI-assisted coding setups today, developers work in what is often described as a "prompt-and-monitor" cycle. Engineers give instructions to an AI system, review its output, and then issue the next prompt. As AI agents become capable of handling more complex tasks, this approach can quickly become overwhelming, especially when multiple agents are involved in different parts of a project.
Cursor's Automations system aims to reduce that constant back-and-forth. Instead of waiting for human instructions, agents can start working when certain conditions are met. For instance, an automation can be triggered when new code is added to the codebase, when a message appears in Slack, or when a scheduled timer activates.
Once triggered, the AI agent can begin tasks such as reviewing code, analysing changes, or checking for issues. Human engineers are then notified only when their input is required.













