Artificial intelligence makes it easier than ever to turn an idea into a working application. Developers can describe a feature, request an implementation, and begin reviewing generated code within minutes. However, this speed creates a temptation to skip planning, requirements analysis, and the critical thinking needed to build reliable software.
This week’s Building Better Developers Weekly Challenge focuses on changing that habit. Following discussions about protecting sensitive information and owning the complete software development lifecycle, Michael Meloche and Rob Broadhead encourage developers to practice using AI as a planning and review partner rather than immediately asking it to generate code.
The challenge is simple: select a small project, create a development plan with AI, and then find ways to improve that plan before implementation begins.
Choose a Small Project or Feature
Start by selecting something you genuinely want to build. It could be a personal utility, an automation script, a small application, or a feature you have wanted to add to an existing project.
Rob shares an example involving a tabletop role-playing game that he used AI to turn into a web-based application. His experience demonstrates how working on a passion project can reveal gaps between what developers imagine and what they actually communicate.
Choose something small enough to understand but interesting enough that you will recognize when AI makes incorrect assumptions.
Ask AI to Create a Plan Before Writing Code
Instead of immediately asking AI to build your application, begin by explaining what you want to accomplish and requesting a development plan.
Ask AI to identify the requirements, major components, potential risks, architecture, testing needs, and implementation steps. Review the result and determine whether it accurately represents your original idea.
Consider including these elements in your initial plan:
- A clear problem statement and project objective.
- Functional requirements and acceptance criteria.
- Proposed architecture and major components.
- Data requirements and security considerations.
- A task list describing implementation steps.
- Testing requirements and expected outcomes.
- Potential risks, limitations, and unanswered questions.
Save the plan in a Markdown document so that it can be reviewed and updated throughout the project.
Challenge AI to Find Problems
Once the initial plan exists, change the role AI plays in the conversation. Instead of asking it to expand the application, have it critique the design.
Michael recommends examining the plan from several perspectives, including security, quality assurance, and user experience. Ask AI what information might be exposed, which requirements appear incomplete, and whether the proposed interfaces meet the intended user’s needs.
Rob adds that developers should pressure-test their designs against realistic failure scenarios. An application may work perfectly for one user but behave differently when hundreds or thousands of people begin using it.
Consider asking questions such as:
- What happens if the application suddenly has 1,000 users?
- What breaks if the database connection disappears?
- How does the application behave without internet access?
- What happens when users submit unexpected information?
- Which security vulnerabilities should be investigated?
- What problems could emerge after a year of normal usage?
These questions help identify weaknesses before they become expensive implementation problems.
Review What AI Missed
The most important part of this exercise is comparing the AI-generated plan with what you originally intended.
Did AI misunderstand the problem? Did it introduce unnecessary functionality or overlook an important requirement? Was the proposed architecture too complicated for the application’s needs?
Sometimes AI makes incorrect assumptions. Other times, the problem is that the developer did not clearly communicate expectations.
Learning to recognize these differences improves both technical planning and the ability to work effectively with AI.
You can also ask another AI model to review the plan, providing an additional perspective. However, the responsibility for evaluating those recommendations remains with the developer.
Complete This Week’s Challenge
Set aside time this week to select a small feature or passion project and take it through a structured planning process with AI.
Create the requirements, review the proposed architecture, identify risks, examine alternative perspectives, and pressure-test the design. Revise the plan based on what you discover, and only then consider moving into implementation.
The objective is not necessarily to finish building the application this week. Success means producing a clearer, more complete plan and understanding which assumptions needed improvement.
AI can generate code quickly, but speed alone does not guarantee quality. By learning to plan before you code with AI, you strengthen the critical-thinking, design, and evaluation skills that remain valuable throughout a developer’s career.
Stay Connected: Join the Developreneur Community
👉 Subscribe to Building Better Developers for more conversations on momentum, leadership, and growth. Whether you’re a seasoned developer or just starting, there’s always room to learn and grow together. Contact us at [email protected] with your questions, feedback, or suggestions for future episodes. Together, let’s continue exploring the exciting world of software development.
Additional Resources
- Why AI Readiness Matters More Than AI Adoption
- AI Implementation Guardrails: How to Scale AI Without Scaling Your Mistakes
- Keeping Forward Momentum When You’re Overloaded: Small Wins + AI Guardrails
- Building Better Developers Podcast Videos – With Bonus Content
WordPress / Yoast SEO
| Focus keyphrase | plan before you code with AI |
| SEO title | Plan Before You Code With AI: Weekly Challenge |
| URL slug | plan-before-you-code-with-ai |
| Category | Weekly Challenge |
| Meta description | Learn to plan before you code with AI. This weekly challenge teaches developers to define requirements, review designs, and test assumptions before building. (157 characters) |
| Excerpt | Before asking AI to build your next application, create a plan. This Weekly Challenge shows how to use AI to define requirements, review architecture, identify risks, and pressure-test an idea. |
Tags (5): AI Development, Prompt Engineering, Software Planning, Critical Thinking, Weekly Challenge,