ditpoo

Sharing best practices I discovered/found for coding using AI based code generation

1. Provide code base/project details and up-to-date context.
2. Separation of concerns in different modules.
3. Create implementation details for each PR, maintain implementation docs.
4. Generate small, focused chunks of code (< 1k lines).
5. Keep functions concise and small.
6. Comment every function; add context before AI generation.
7. Keep those comments updated with every change on that code

ditpoo

Untested but Promising Blueprint for Writing Production-Ready Code with AI Code Generation

What's Lacking / Required:

• Defined processes, practices, and workflows.
• The ability to generate working or easy to debug/fix code.
• The ability to generate working or fixable production-ready code.
• Efficient management of code files with thousands of lines of code (loc).
• Management of file/code history, changes, and context.

Goals:

ditpoo

scene: POV from eagle's head flying over 18th-century Paris sky during French Revolution — revolution banners, crowded alleys, crumbling rooftops, with wing shadows and occasional feather flicks visible, eagle dives past Notre-Dame, banks over rioting streets, weaves between cannon towers, final dive over revolution square.

aespect_ratio: 2.35:1

scene_tone: epic, historic, fluid

shot_type: ultra-wide aerial shot, with motion blur trails.
camera_movement: swooping dives, rooftop skims, banking turns around towers and smoke columns

lighting_effects: dramatic sunset with rim light on rooftops, warm haze through smoke,

ditpoo

# GPT-5 Image Gen Prompt Template Example.

Capture a scene with **[subject(s)]** over **[location and environment]**. **[brief narrative beat or moment to capture]**. In a style of **[art/style family]** rendered in **[pipeline/medium]**, with **[hallmark style traits]**.

## Non-negotiables
<series_constants>

<Character bible>  
- **[Character A]** — [build/skin/face markers/hair/defining feature(s)].