Prompting

To ensure that you get the most accurate and helpful responses from Credal, it’s crucial to craft your prompts effectively.

Here’s why prompt engineering is worth your attention:

1. Improves Accuracy: Well-crafted prompts lead to more precise and relevant answers, reducing the likelihood of misinterpretations or general responses.
2. Saves Time: By clearly directing the AI, you can obtain the desired information with fewer attempts.
3. Enhances Complexity Handling: Complex tasks often require nuanced understanding. Good prompts translate intricate questions into a form that the AI can process effectively.
4. Facilitates Innovation: Mastering prompt engineering allows you to push the boundaries of AI’s applications, driving innovation and enabling new solutions to complex problems.

This guide will provide you with the foundational understanding and practical steps to optimize your interactions with Credal for precise and valuable responses.

1. What is the Background Prompt

This is arguably the most effective way to customize your copilot. It defines the landscape in which your copilot operates, detailing its role, response style, and the types of queries it should handle. Crafting an effective background prompt ensures your copilot can seamlessly integrate into your specific use case.

2. Constructing a Background Prompt

Role and Goal

Outline the specific role your copilot is expected to fulfill. Include details such as:

• Who: The specific team or individual utilizing the copilot.
• What: The usual nature of queries or tasks it will address.
• Where: The operational context, be it customer service, technical support, etc.

Example

Role and Goal:
You assist the financial advisory team by analyzing client interactions to enhance service quality. You focus on assessing customer sentiment and identifying potential upsell opportunities from each interaction. Provide insights on key financial topics discussed.

Output Format

LLM’s can be a bit unpredictable, but luckily we can just tell them what to do with clear instructions. Providing a template for expected responses is an example of this.

Here’s an example of setting up an output format for entity extraction:

Extract the important entities mentioned in the context.
Output format:
Company names: <comma_separated_list_of_company_names>
People names: <comma_separated_list_of_people_names>
Specific topics: <comma_separated_list_of_specific_topics>
General themes: <comma_separated_list_of_general_themes>

Another more crude example:

Provide a riddle using the following structure:
- Riddle: <the puzzle or brainteaser>
- Answer: <the solution to the puzzle>
- Difficulty level: <a difficulty rating from 1 (easy) to 10 (challenging)>

Response Guidelines

Craft detailed instructions that anticipate potential pitfalls and standard operational procedures. Your Copilot will only know as much as you tell it. Noticing that the responses are too long? Specificy response length! Think there’s too much fluff in the language? Ask for concise language.

Credal has prompt snippets right under the background prompt in the Copilot Configuration tab. We recommend using those as a starting point for generating more useful responses, however your specific use case may require much more detail!

Here’s an example of a fully crafted background prompt…

Role and Goal:
You work in Credal's Customer Success team, and your goal is to enhance customer support quality by analyzing interactions for performance improvement and coaching. For each CS interaction transcript, you will provide a score per category and actionable feedback per review tag based on the pinned QA Scorecard.
Output:
1 scoring table with following 2 columns: Category, # of Points
1 feedback table with following 2 columns: Review Tag, Actionable Feedback
Response Guidelines:
-If there are any trust-related issues in the transcript based on the trust-specific review tags in the scorecard, auto-fail the entire ticket and give an overall score of 0; if there are no trust-related issues, put "N/A"
-Scorecard categories have specific corresponding review tags. Give full points in this category only if there are no review tag failures, and 0 points if the transcript contains any of the corresponding review tags
-"# of Points" refers to score per category for the attached transcript
-Include all review tags in the table and each review tag (e.g., "[T] Disclosing private Credal/TaskUs Information e.g. leaking product launches") should be its own row with corresponding feedback
-Make the feedback specific, pull examples from the transcript, and provide sample improved talking points
-Your tone is professional, concise, helpful, and coaching
--- Start Context ---
{{data}}
--- End Context ---

3. Start with Simple Prompts and Iterate

Begin with simple prompts (zero-shot) and evolve to more complex ones (few-shot) as needed. If further precision is required, consider fine-tuning the AI model:

Zero-shot Example:

List key features of the following product.
Text: {text}
Features:

Few-shot Example:

Text 1: "The new Credal AI platform offers enhanced data security."
Features 1: enhanced data security
##
Text 2: "Credal AI enables seamless API integration."
Features 2: seamless API integration
##
Text 3: {text}
Features 3:

Another example…

• Zero-shot Example:

  Product Description: "This is a state-of-the-art vacuum cleaner with HEPA filters."
  List its selling points.

• Few-shot Example:

  Text 1: "This oven offers precise temperature control and a self-cleaning feature."
  Selling Points 1: precise temperature control, self-cleaning feature
  ##
  Text 2: "This smartphone has a high-resolution camera and long-lasting battery."
  Selling Points 2: high-resolution camera, long-lasting battery
  ##
  Text 3: "This laptop includes a high-speed processor and lightweight design."
  Selling Points 3:

4. Clarity Over Vagueness

Avoid imprecise language. Precise prompts lead to more accurate outcomes:

Less effective:

The description for this product should be fairly short, a few sentences only.

Better:

Describe this product in a 3 to 5 sentence paragraph.

5. Say What to Do Instead of What Not to Do

Provide positive instructions to guide the model effectively.

Less effective:

DO NOT REPEAT.

Better:

Encourage concise problem-solving with clear instructions. Instead of asking the same question, guide the user to review specific troubleshooting steps or the FAQ at www.example.com/FAQ.

6. Save your work in a “Suggested Question”

So, you did all the work to craft the perfect prompt for your super specific use-case… only to lose it when you hit send? Think again. If you’re a collaborator on a copilot, you can copy-paste your work of art into a Suggested Question in the Copilot config. Now when you log into the webUI, you can reuse your prompt with the click of a button!

We currently don’t support autocompleting prompts for our Slack integration. For Copilots deployed to Slack, we recommend crafting the background prompt to handle the diverse types of messages sent in the channel of your choice.

7. Use the Latest Model

Utilize the most recent and capable models to achieve the best results. Newer models are generally more adept at understanding and following your prompts.

8. Incorporate Feedback Loops

Continuously refine your prompts based on your impression of Copilot responses to improve accuracy and relevance. Negative feedback logs are a great resource for this!

We are currently working on some features that will incorporate negative feedback and usage trends into the copilot configuration workflow by suggesting ways in which you can improve your background prompt. Stay tuned!

9. What not to ask…

1. Anything about access controls. That’s on us, we double check access policies on the Credal side before sending anything to the LLM!
2. Anything about a Credal concept (Copilots, Document collections, “pinned” sources, etc.). The LLM doesn’t know what these are, we will do the searching and consolidation of information for you.

Feel free to adapt these examples and practices to best fit your specific use cases with Credal AI.