In this episode, I am joined by Ras Mic, a full stack engineer & YouTuber, where we dive deep into the frameworks and strategies on how to best use Cursor AI. Mic shares his unique insights into how to use and set up Cursor to make the experience of building on top of Cursor as easy and seamless as possible. Learn how to use Cursor like a pro!
1) Plan before coding • Sketch out ideas (Figma, iPad, paper) • Use V0 for mockups (10-15 prompts minimum) • Set your AI assistant up for success "Half the game is acting like a developer" - Mike
2) Use cursor.directory for better prompts • Copy tech-specific prompts • Create .cursor_rules file in project root • Drastically improves Cursor's output Pro tip: Customize prompts for your exact tech stack
3) Tag relevant documentation • Add official docs to Cursor (Next.js, Supabase, etc.) • Ensures up-to-date, accurate responses • Cursor reads docs when answering queries
4) Ask other AI models when stuck • Copy errors to Claude, GPT-4, etc. • Include failed solutions for better context • Different models can solve problems others can't
5) Use AI to explain code • Great for learning & documentation • Ask to explain like you're a beginner • Follow up on confusing concepts "The more you do this, the more you learn patterns" - Mike
6) Leverage existing code templates • Start with boilerplate (auth, DB, payments) • Build on top of proven foundations • Saves time on repetitive tasks
7) Add comments to your code • Ask AI to explain & document your code • Improves readability & understanding • Great for team collaboration
8) Duplicate & modify existing functionality • Copy working code from other parts of your project • Ask AI to adapt it for new use cases • Provides context for better results Key takeaway: Give AI as much context as possible for best results!
Want more free ideas? I collect the best ideas from the pod and give them to you for free in a database. Most of them cost $0 to start (my fav)
Episode Timestamps: 0:00 Intro 02:20 Tip 1: Plan before coding 11:11 Tip 2: Use cursor.directory for better prompts 19:28 Tip 3: Tag relevant documentation 26:12 Tip 4: Ask other AI models when stuck 29:40 Tip 5: Use AI to explain code 35:22 Tip 6: Duplicate & modify existing functionality 36:24 Tip 7: Leverage existing code templates
