Overview

Note: Vly is currently free to use! These best practices will help you get better results right now, and they'll also be valuable when we introduce usage-based pricing in the future.

Following these best practices will help you build better applications faster, get more accurate results from the AI agent, and create maintainable, scalable projects.

Core principles

Start with clear goals

Before beginning any project:

• Define the primary purpose of your application
• Identify key user flows and functionality
• Consider the target audience and their needs
• Plan the main features before diving into details

Build incrementally

Rather than requesting everything at once:

• Start with core functionality and essential features
• Add complexity gradually as the foundation is solid
• Test each major addition before moving to the next feature
• Create checkpoints at significant milestones

Communicate effectively

Clear communication with the AI agent leads to better results:

• Be specific and descriptive in your requests
• Provide context and reasoning for your decisions
• Reference existing components when making modifications
• Give feedback quickly when something isn't right

Efficiency guidelines

Communication optimization

Get faster, more accurate results from your AI interactions:

• Bundle related requests into single prompts when possible
• Avoid repetitive explanations by building on previous context
• Use precise language to get to the point quickly
• Reference previous work instead of re-explaining requirements

Prompt optimization

Structure your requests for maximum clarity:

• Lead with the main request before providing details
• Use examples to illustrate complex requirements
• Break down multi-step processes into clear sequences
• Specify constraints and preferences upfront

Resource management

Work efficiently with project resources:

• Upload assets early if you know you'll need them
• Organize files logically as your project grows
• Remove unused code periodically to maintain clarity
• Document complex logic for future reference

Quality assurance

Testing approach

Ensure your application works reliably:

• Test core functionality after major changes
• Check responsive design on different screen sizes
• Verify user flows from start to finish
• Test edge cases and error scenarios

Code organization

Maintain clean, understandable code:

• Follow consistent naming conventions for components and files
• Group related functionality in logical structures
• Keep components focused on single responsibilities
• Maintain clear separation between frontend and backend logic

Performance considerations

Build applications that perform well:

• Optimize images and assets for web delivery
• Use appropriate data structures for database queries
• Implement proper loading states for better user experience
• Consider caching strategies for frequently accessed data

Collaboration best practices

Team coordination

When working with multiple people:

• Establish clear roles and responsibilities
• Communicate changes that affect shared components
• Create checkpoints before major collaborative work
• Use descriptive commit messages for version history

Project organization

Structure projects for team success:

• Document architectural decisions in your prompts
• Maintain consistent code style across the project
• Use meaningful component and file names
• Keep related functionality grouped together

Common pitfalls to avoid

Overcomplicating requests

Avoid these common mistakes:

• Don't request too many features in a single prompt
• Avoid vague or ambiguous language
• Don't assume the AI knows your unstated preferences
• Avoid changing direction midway through complex requests

Technical debt

Prevent future maintenance issues:

• Don't ignore warnings or error messages
• Avoid hardcoding values that should be configurable
• Don't skip proper error handling for user interactions
• Avoid accumulating unused code and assets

Inefficient workflows

Common efficiency problems:

• Starting over instead of iterating on existing work
• Not leveraging checkpoints for experimentation
• Forgetting to test changes before adding new features
• Not organizing assets and files systematically

Getting the most from vly.ai

Leverage AI capabilities

Take advantage of what vly.ai does best:

• Let the AI make architectural decisions based on best practices
• Use the AI's knowledge of modern web development patterns
• Trust the AI's integration of different technologies
• Ask for explanations when you want to understand the implementation

Provide effective feedback

Help the AI agent improve results:

• Be specific about what works and what doesn't
• Explain your preferences for future similar requests
• Point out inconsistencies in styling or functionality
• Suggest improvements based on user experience

Continuous improvement

Refine your approach over time:

• Learn from successful prompts and reuse effective patterns
• Keep notes on what works well for your project types
• Experiment with different communication styles
• Share insights with team members

Next steps

Dive deeper into specific best practices:

• Optimize your AI interactions
• Prompt efficiency

These practices will evolve as you gain more experience with vly.ai. The key is to start with these fundamentals and adapt them to your specific needs and project requirements.