Skip to main content

MWD Methodology

Microwise Development (MWD) is a development methodology optimized for AI-assisted coding.

Core Principles​

1. Micro Commits​

Make small, focused commits that do one thing well.

# Good: Specific, focused
git commit -m "Add email validation to signup form"

# Bad: Too broad
git commit -m "Update user module"

2. Incremental Progress​

Break large tasks into small, verifiable steps:

Task: Add user authentication
  ├── Step 1: Create User model
  ├── Step 2: Add password hashing
  ├── Step 3: Create login endpoint
  ├── Step 4: Add session management
  └── Step 5: Write tests

3. Clear Task Boundaries​

Each AI session focuses on one well-defined task.

Why MWD?​

Traditional development workflows don't account for AI limitations:

TraditionalMWD
Large commitsMicro commits
Long sessionsFocused sessions
Complex contextClear boundaries
Manual trackingAutomated logging

Using MWD Profile​

Start a session with MWD profile:

ilogsession start my-task -p mwd

The MWD profile configures the AI to:

  • Make frequent, small commits
  • Document progress incrementally
  • Ask for clarification on ambiguous tasks
  • Test changes before committing

Best Practices​

1. One Task Per Session​

# Good: Focused session
ilogsession start add-email-validation -p mwd

# Avoid: Multiple unrelated tasks in one session

2. Commit Often​

After each logical change, commit:

git add .
git commit -m "Add email format validation"
# Continue to next step...

3. Document As You Go​

The AI will automatically document:

  • What was changed
  • Why it was changed
  • How to test it

Example Workflow​

# 1. Start session
ilogsession start feature-x -p mwd

# 2. AI works on the task, making micro commits
# 3. Session logs are saved automatically
# 4. If session restarts, context is preserved

# Check session logs
ilogsession logs feature-x