Logo Lanfrica

vic-322/The-Precision-Prompting-Challenge

Domain:

healthcarenatural language processing
Creator:
vic
Host:
You are the newly appointed AI Strategy Lead at AfyaTech, a Nairobi-based health tech startup. Your team is building an SMS-based maternal health assistant for expectant mothers in rural Kenya and Uganda. Early testing shows your current AI prompts produce generic, urban-centric advice that ignores: The-Precision-Prompting-Challenge You are the newly appointed AI Strategy Lead at AfyaTech, a Nairobi-based health tech startup. Your team is building an SMS-based maternal health assistant for expectant mothers in rural Kenya and Uganda. Early testing shows your current AI prompts produce generic, urban-centric advice that ignores: Prompt A: Nutrition Advice (Localized) AIM Framework A (Audience): Pregnant women in rural Kenya and Uganda with limited income and access to markets I (Intent): Provide practical, affordable, culturally relevant nutrition advice M (Mode): SMS-friendly, simple language, actionable tips MAP Framework M (Model Constraints): Must use locally available foods such as: ugali, sukuma wiki, matooke, beans, groundnuts); avoid imported foods A (Augmented Context): Consider seasonal food availability, local markets, and common diets P (Prompt Instruction): “Provide 3–5 simple nutrition tips for a pregnant woman living in a rural area of Kenya or Uganda. Use locally available foods such as ugali, sukuma wiki, matooke, beans, groundnuts, milk, and eggs. Ensure tips are affordable, culturally appropriate, and easy to follow via SMS. Avoid suggesting foods that may not be locally available.” Key Improvement: Reduces hallucination and irrelevance by anchoring advice in real, local food systems instead of generic global recommendations. Prompt B: Appointment Reminders (Context-Aware) AIM Framework A (Audience): Pregnant women in rural areas with limited transport and clinic access I (Intent): Encourage timely clinic visits without causing stress or impractical expectations M (Mode): Short SMS reminders with flexible planning MAP Framework M (Model Constraints): Account for travel time, transport costs (e.g., motorbike, walking), clinic days, and CHW availability A (Augmented Context): Rural clinics may operate on specific days; community health workers (CHWs) may assist locally P (Prompt Instruction): “Generate a friendly SMS reminder for a pre …

Languages