Architecting the Architect: Using AI to Instruct AI to Build with AI

Explore how AI is revolutionizing software development as we use ChatGPT to guide Windsurf IDE’s SWE-1 model in building a complete expense-sharing application. This series demonstrates the future of AI-assisted development, from prompt engineering to deployment. ChatGPT Instructor

🚀 The Future of Software Development is Here

In recent years, AI has evolved from a coding assistant to a full-stack developer. This series demonstrates how to leverage ChatGPT (GPT-4o with Deep Research) to instruct Windsurf IDE (SWE-1 model) in building a complete, production-ready application—without writing a single line of code yourself.

“The most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it.” — Mark Weiser

📚 What We’ll Be Building: Splitwisely

I decided to build Splitwisely, an expense-sharing application similar to Splitwise that helps friends and roommates track and settle shared expenses. I chose this project because it covers many aspects of modern app development:

User authentication and management
Group creation and management
Expense tracking and splitting
Real-time balance calculations
Clean, responsive UI

🛠️ Series Overview

Part 1: Setting Up ChatGPT to ask right questions

Crafting optimized prompts for Windsurf IDE
Establishing clear project intentions
Creating an effective AI workflow

Part 2: Mastering AI-to-AI Communication

Techniques for refining prompts
Best practices for AI collaboration
Ensuring clarity in AI instructions

Part 3: Defining Project Requirements

Using AI to ask the right questions
Creating comprehensive project specifications
Validating requirements before implementation

Part 4: Configuring Windsurf IDE

Setting up development environments
Establishing coding standards
Creating reusable templates

Part 5: From Idea to Implementation

Generating Product Requirements Documents (PRD)
Breaking down features into tasks
Creating actionable development plans

Part 6: AI-Powered Development

Implementing features with AI assistance
Code review and iteration
Testing and quality assurance

Part 7: Documentation and Maintenance

Automated documentation generation
Keeping documentation in sync
Long-term maintenance strategies

🔍 Why This Matters

The traditional software development lifecycle is being transformed by AI. This series demonstrates how:

Productivity - Reduce development time from weeks to days
Accessibility - Enable non-coders to build complex applications
Quality - Leverage AI’s ability to follow best practices
Innovation - Focus on what matters while AI handles implementation

🎯 Who Should Follow This Series

Developers looking to enhance their workflow with AI
Product Managers wanting to prototype ideas quickly
Startup Founders with limited technical resources
AI Enthusiasts curious about practical applications

🚀 What’s Next?

In Part 1, we’ll dive into setting up ChatGPT to effectively communicate with Windsurf IDE. You’ll learn how to craft prompts that produce high-quality, production-ready code.

Ready to see the future of software development? Let’s begin!

References

The Ninja Studio. (2025). AI-Assisted Software Development in 2025. https://www.theninjastudio.com/blog/ai-assisted-software-development
DevOps.com. (2025). Windsurf Launches SWE-1: AI Models Built for the Entire Software Engineering Process. https://devops.com/windsurf-launches-swe-1-ai-models-built-for-the-entire-software-engineering-process/