1. Introduction2. Team Structure
3. Process Flow
3.1. Module Learning & Experience Design3.2. Individual Flow Design3.3. Narrative Design & Skills Mapping3.4. New Assets Creation3.5. Annex: World Building - Learning & Experience Design3.6. Annex: World Building - Story Design3.7. Process Flow Conclusion4. Caveats
4. Caveats5. LLM Prompt Engineering Techniques
5.1. Use The Latest Model5.2. Zero-shot Prompting5.3. Few-shot Prompting5.4. Chain-Of-Thought Prompting5.5. Structuring Prompts5.6. Describing Prompts5.7. Editing Prompts5.8. Extending Responses5.9. Multiple Users Collaborating6. Text-to-Image Prompting Engineering Techniques
6.1. General Techniques6.2. Photography6.3. Architecture6.4. Various Aesthetic Styles6.5. Product & Material5.7. Editing Prompts
While using ChatGPT, it can sometimes be more effective to edit an existing prompt (clicking on the edit button beside the prompt) rather than creating a new one.
- Context Preservation: Editing an existing prompt retains the context and allows us to fine-tune our query or instruction, which may lead to a more accurate and relevant response from the AI.
- Improved Clarity: By refining the original prompt, we can clarify any ambiguities or vagueness that might have caused the AI to generate an inaccurate or off-topic response.
Whenever we want to make minor edits while retaining the context of the prompt, we use the edit function.
If we want to make a drastic change, we simply write a new prompt.
← Previous
Next →
On this page: