/prɒmpt ˈdrɪvən dɪˈvɛləpmənt/
Definition
A cutting-edge software methodology in which engineers outsource thinking to large language models, then spend hours refining prompts instead of code. The process optimises for syntax correctness, semantic confusion, and creative self-deception.
Common Manifestations
- Developers asking, “Can you make this more enterprise?” to a chatbot.
- Repositories containing README files longer than the actual codebase.
- Meetings debating whether “as a senior developer” should be part of the prompt.
- Auto-generated functions with names like do_the_thing_but_better_final_v2().
- Entire teams rebranding prompt engineering as “AI Pair Programming.”
Usage Example
“We shipped an MVP entirely through Prompt-Driven Development — no one knows how it works, but the AI said it looks good.”
HR Guidance
PDD fosters creativity, speed, and plausible deniability.
Encourage teams to document their prompts rather than their reasoning.
If the code fails, celebrate the journey — after all, it’s about prompting progress, not achieving results.