Organization-level Context for AI
What is organization-level context for AI?
Organization-level context allows you to provide global, programmable instructions to the AI. It guides the AI to adhere to your company’s specific business logic, terminology, and data standards.
Organization-level context typically includes:
- Business Background: High-level information about your company’s industry and goals.
- Operational Knowledge: Unique processes and industry nuances not yet captured in your AMQL formulas.
- Response Preferences: Standardized instructions for the AI’s tone, language, and writing style.
- Data Exploration: Configurations that map specific user roles or attributes to their most relevant datasets.
How to add this context?
In Holistics, you can add your organization-level context in the file context.aml. We have already created an empty file for you to start with.
You can find it under either:
- Development > Project Explorer > settings > ai
- Your Tenant name > Organization Settings > More Settings > AI Settings
Next step, define your logics by AML or natural language under const ai_context = @md.
Examples
Add business background
const ai_context = @md
## Business Background:
Adventure Works sells high-quality road bikes and mountain bikes,
and cycling accessories such as bike racks, stands, tires and tubes, etc.
;;
Add AI repsonse preference
const ai_context = @md
## AI Response Preferences
- Be concise, professional and objective
- Always give 1-2 key insights and suggest next steps for deeper analysis
;;
Customize responses by user attribute
const product_context = '''
Preferred dataset: ${products_dataset}
'''
const customer_context = '''
Preferred dataset: ${customers_dataset}
'''
const purchasing_context = '''
Preferred dataset: ${purching_dataset}
'''
const ai_context = @md
${
if (H.current_user.team == 'Business Analyst') {
product_context
} else if (H.current_user.team == 'Growth & Marketing') {
customer_context
} else if (H.current_user.team == 'Purchasing'){
purchasing_context
}
}
;;