Hacker News is a community-driven platform for sharing and discussing technology news, startups, and programming-related topics. Through user submissions and comments, Hacker News offers insights into emerging technology trends, industry developments, and entrepreneurial ventures. Readers can participate in discussions, share their insights, and stay informed about the latest advancements in technology and innovation.

Hacker News

Google Antigravity is a new AI-powered IDE featuring tab autocompletion, natural language code commands, and context-aware agents. The platform offers synchronized agentic control across editor, terminal, and browser, with a task-based approach to monitoring agent activity. It supports managing multiple agents simultaneously from a central mission control view and is available at no charge for developers.

Google Antigravity

Amit yadav

I’ve tried it. After a first request on a relatively small codebase, I got a message with the Gemini model that I reached the credit limits. Then changed to Sonnet, and I was able to work with. While it promised that it can test the issues in the browser, it was no usable. I had an issue that a TOC on the page not scrolled to the relevant section in the text on clicking it. I worked more than an hour on it, stated several times it tested it and it is working. Then I checked it manually, told him what is the issue, and it fixed it. Next request, and I reached the credit limits again.
The concept looks good, but the execution needs more fine tuning. My experience is that some competitors are much more smarter than this, even without visual in-browser testing. And no, it is not Claude, while that works much better, the frequent waits for compressing the conversation is crazy.

Works fine for software testing purpose. It generated working playwright solution with 2 UI and 1 api tests after just one prompt :). Looks promising