Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

HPCC-32862 Add Copilot article to devdoc #19229

Open
wants to merge 1 commit into
base: candidate-9.8.x
Choose a base branch
from
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
84 changes: 84 additions & 0 deletions devdoc/userdoc/copilot/CopilotPromptTips.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,84 @@
# Copilot Prompt Tips
Unlock the full potential of GitHub Copilot with these carefully curated prompts designed to streamline your workflow and enhance productivity.

Whether you're summarizing complex texts, brainstorming innovative ideas, or creating detailed guides, these prompts will help you get the most out of Copilot's capabilities.

Dive in and discover how these simple yet powerful prompts can save you time and effort in your daily tasks.

## Generic Prompts

Here are a few simple prompts that can save time:

- Provide a brief summary of the text below. Show the key points in a bullet list.

- List [N (number)] ways to [accomplish a specific goal or solve a problem]? Include a short description for each approach in a bullet list.

- **Example**: List 10 ways to reduce redundancy in text. Include a short description for each approach in a bullet list.

- What are the key differences/similarities between [concept A] and [concept B]? Present the information in a table format.

- **Example:** What are the key differences/similarities between a quicksort ad a bubble sort? Present the information in a table format.

- Explain [complex topic] in simple terms. Use analogies or examples to help make it easier to understand.

- **Example:** Explain generative AI in simple terms. Use analogies or examples to help make it easier to understand.

- Brainstorm [N (number)] ideas for [a specific project or topic]? Include a short description for each idea in a bullet list.

- **Example:** Brainstorm 7 ideas for an instructional video for new programmers? Include a short description for each idea in a bullet list.

- Create a template for [a specific type of document, such as a business email, proposal, etc.]. Include the key elements to include in a bullet list.

- **Example:** Create a template for product version release announcement. Include the key elements to include in a bullet list.

- Write a step-by-step guide on how to [specific task or procedure]. Number the steps to improve clarity.

- **Example:** Write a step-by-step guide on how to make a PB & J sandwich. Number the steps to improve clarity.

## Specific Prompts

These prompts are more focused and are meant to accomplish a specific purpose. They are included here to spark your imagination.

If you think of others that could be included here to share with the world, please send your ideas to <[email protected]>.

- Write comments for this ECL code in javadoc format

- Create an ECL File Definition, record structure, and inline dataset with [N (number)] of records with the following fields [list of fields]

- **Example:** Create an ECL file definition with a record structure and an inline dataset with 20 records for a dataset of animals with the following fields: ReferenceNumber, AnimalName, ClassName, FamilyName, Color, Size, and LifeExpectancy.

- Write ECL code to classify records into [categories].

- **Example:** Write ECL code to classify customer feedback into neutral, negative, or positive categories.


## How to Avoid AI Hallucinations with Good Prompts
AI hallucinations refer to instances where artificial intelligence systems generate information or responses that are incorrect, misleading, or entirely fabricated.

Hallucinations can occur due to various reasons, such as limitations in the training data, inherent biases, or the AI's attempt to provide an answer even when it lacks sufficient context or knowledge.

Understanding and mitigating AI hallucinations is crucial for ensuring the reliability and accuracy of AI-driven applications.

Creating effective prompts is essential to minimize AI hallucinations. Here are some tips to help you craft prompts that lead to accurate and reliable responses:

- **Be Specific and Clear**: Ambiguous prompts can lead to incorrect or irrelevant answers. Clearly define the task and provide specific instructions.

- **Example**: Instead of asking "What is AI?", ask "Explain the concept of artificial intelligence and its primary applications in healthcare."

- **Provide Context**: Give the AI enough background information to understand the task. This helps in generating more accurate responses.

- **Example**: "Given the following text about climate change, summarize the key points in a bullet list."

- **Use Constraints**: Limit the scope of the response by specifying constraints such as word count, format, or specific details to include.

- **Example**: "List 5 benefits of renewable energy in a bullet list, each point not exceeding 20 words."

- **Ask for Evidence or Sources**: Encourage the AI to provide evidence or cite sources for the information it generates.

- **Example**: "Explain the impact of social media on mental health and provide references to recent studies."

- **Iterative Refinement**: Start with a broad prompt and refine it based on the initial responses to get more accurate results.

- **Example**: Begin with "Describe the process of photosynthesis." If the response is too vague, refine it to "Describe the process of photosynthesis in plants, including the role of chlorophyll and sunlight."

By following these guidelines, you can reduce the likelihood of AI hallucinations and ensure that the responses generated are accurate and useful.
Loading