🚀 Level up your mobile dev skills at https://codewithbeto.dev/learn! Use code HACKTHEPRICE25 for 25% OFF!
📱AI Tattoo App → https://codewithbeto.dev/resources/aiTattoo

In this video, we take Cursor 2.0 for a spin — testing out Composer 1, the new AI model for coding introduced in this release. We refactor a real-world payments screen and compare how different models like Sonnet 4.5, GPT-5 Codex  and Composer handle the update.

Master React Native alongside me → https://codewithbeto.dev/learn

👨🏻‍💻  🌟  Be the mobile expert on your team 🌟  👨🏻‍💻
Learn mobile development with courses, templates, and full-stack codebases built to help you ship better apps: 

- React Native Course → https://codewithbeto.dev/learn
- React with TypeScript Course → https://codewithbeto.dev/learnReact
- Git & GitHub Course → https://codewithbeto.dev/learnGit

For other ways to support my work, please consider:

- GitHub Sponsors: https://github.com/sponsors/betomoedano
- Become a Code with Beto channel member: https://www.youtube.com/channel/UCh247h68vszOMA_OWpGEa5g/join

⌚️  Timestamps ⌚️
0:00 — Intro
1:19 — Announcement
2:11 — Project overview
3:32 — Agent tab overview
5:08 — Adding images as context
6:06 — Crafting a prompt
8:18 — Submitting prompt to 3 agents
9:04 — Reviewing composer changes
12:06 — Reviewing GPT-5 Codex changes
12:40 — Reviewing Sonnet 4.5 changes
13:18 — Quick review of all models
14:40 — Recap

⭐️   If you want to learn more about me, check this links: 

Code With Beto → https://codewithbeto.dev
🦋 → https://bsky.app/profile/codewithbeto.dev
X → https://x.com/betomoedano
Github → https://github.com/betomoedano
Linkedin → https://www.linkedin.com/in/betomoedano/
Discord → https://discord.com/invite/dbYfWFw862
Medium → https://medium.com/@betomoedano01
Figma → https://www.figma.com/@codewithbeto

#reactnative #WhisperAI #speechtotext

Code with Beto

A hands-on walkthrough of Cursor 2.0's new features, demonstrated through a real React Native mobile app project. Key highlights include the new Composer AI model (fast, built in-house), and the multi-agent tab that lets you run the same prompt across up to four different models simultaneously (Composer, Claude Sonnet 4.5, GPT-5 Codex) and compare results. The demo involves revamping a paywall UI, with Sonnet 4.5 ultimately declared the winner despite being slower, while Composer is praised for speed. Practical tips on using image context, file references, and iterative prompting are shown throughout.

Cursor 2.0: Crash Course for Beginners