In this episode, I am joined byJason Zhou, product manager & YouTuber, where we dive deep into AI coding fundamentals. Jason shares his unique insights into how to use and set up AI to streamline to the process of building you next app. Learn how to use AI like a pro!
Timestamps:
00:00 Intro
00:20 Preview of Jason’s Workflow
07:27 Case Study: Building a Reddit Analytics Platform
08:56 Writing detailed specs upfront
11:40 Core Functionality Breakdown
12:51 Researching Packages for Implementation
16:13 Debugging and Iteration Process
19:04 Demonstrating the Reddit Post Fetching
19:59 Categorizing Reddit Posts with OpenAI
24:29 Debugging Continued
29:13 Using Step-by-Step Debugging Techniques
32:15 Setting Up the Project Structure
35:22 Generating Detailed Documentation
38:55 Building the Application with Cursor
46:01 Transitioning to UI Improvements
50:00 Exploring Startup Ideas
51:58 Implementing Reusable Authentication Prompts
56:04 Conclusion: Structuring AI Coding for Success
57:53 Where to Find More Resources and Tutorials
1) The problem: Large language model coding is unpredictable. Tutorials make it look easy, but reality is different. Jason’s solution? A structured workflow for more consistent results.
2) Step 1: Write detailed specs upfront
– Simulate how product teams work in reality
– Uncover uncertainties early
– Include project overview, core functionality, and implementation docs
3) Step 2: Use the right tools for the job
– Cursor for functionality
– V0 for UI touchups
– This order leads to more success than trying everything at once
4) Step 3: Develop reusable modular prompts
– Create a collection of standard functionalities (e.g., user auth, payment)
– Share these prompts within the community
– Speeds up development process significantly
5) Pro tip: When writing specs, think like a product manager AND an engineering manager. Use tools like ChatGPT to help fill in the gaps and create more detailed PRDs.
6) Debugging hack: When Cursor gives vague errors, ask it to "add debug info to the code". This gives you more context to communicate the issue effectively.
7) UI design strategy:
– Build functionality first
– Use V0 for UI improvements later
– Provide style examples (e.g., "similar to Vercel") for better results
8) Jason’s secret weapon: Modular prompts for common features
– Example: Auth setup prompt that works across projects
– Saves time and ensures consistency
9) The future of AI coding?
– Aggregators for modular prompts
– New ways to integrate prompts into developer tools
– More streamlined workflows (like Bolt) to automate manual steps
10) Remember: This approach might feel more structured than expected, but compare it to traditional development – it’s still a massive improvement!
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)
Get access: https://www.gregisenberg.com/30startupideas
Work with me and my team: LCA — world’s best product design firm to build apps, websites and brands people love. https://latecheckout.agency
BoringAds — ads agency that will build you profitable ad campaigns http://boringads.com/
BoringMarketing — SEO agency and tools to get your organic customers http://boringmarketing.com/
Community Empire – a membership for builders who want to build cash-flowing businesses http://communityempire.co/
FIND ME ON SOCIAL
X/Twitter: https://twitter.com/gregisenberg
Instagram: https://instagram.com/gregisenberg/
LinkedIn: https://www.linkedin.com/in/gisenberg/
FIND JASON ON SOCIAL
X/Twitter: https://x.com/jasonzhou1993
Youtube: https://www.youtube.com/@AIJasonZ
