Asking AI health questions effectively means framing them with enough context to get specific answers — including your health history, current situation, and what you have already tried — rather than asking broad questions that produce broad answers. This concept covers health prompt design as a practical skill for getting useful information from AI health tools.
Prompt engineering is the skill of asking AI questions in a way that gets better answers. A vague prompt ("Design me a workout") gets a generic answer. A specific prompt ("Design a 4-day upper/lower split for someone with 2 years lifting experience, recovering from a shoulder injury, who has 1 hour per session and wants to build strength in bench and deadlift") gets a targeted answer tailored to your situation.
This matters because AI is only as good as the information it receives. Garbage in, garbage out. But excellent prompts in create excellent recommendations out.
Start with context: Who are you? Age, current fitness level, training experience, goals. "I'm a 35-year-old with 3 years of lifting experience, trying to build muscle while losing fat."
Add constraints: What are your limitations? Time available, equipment access, injuries, preferences. "I have 45 minutes per session, 4 days available, I have access to a full gym, and I prefer not doing cardio."
Be specific about the goal: Not just "get fit," but what that means to you. "I want to increase my barbell bench press by 20 pounds in 8 weeks while staying at my current bodyweight."
Ask for specific format: Do you want a week-by-week plan, daily breakdown, list of principles? "Give me a day-by-day workout schedule for the next 4 weeks."
Request reasoning: Ask why, not just what. "For each week, explain why you're prioritizing that training split and what adaptation you're targeting."
Poor: "Give me a nutrition plan."
Expected result: Generic 2,000-calorie template that fits nobody specifically.
Good: "I'm 165 pounds, female, doing 4 days of strength training weekly with 2 active recovery days. I want to build muscle without excessive fat gain. Design a daily nutrition plan with specific macros and meal timing around my training."
Expected result: Personalized calorie/macro targets with meal suggestions timed to your workouts.
Good prompt engineering doesn't end with one question. After getting an answer, ask clarifications: "Why did you structure the workout that way?" "What would change if I only had 30 minutes instead of 45?" "How would I progress after 4 weeks?" Each follow-up refines the answer closer to your actual needs.
Asking AI to diagnose medical conditions (it can't). Assuming generic answers are personalized for you. Not providing your actual constraints. Accepting the first answer without follow-up questions.
Try this: Take a fitness or nutrition question you have. First, ask it vaguely to an AI tool (ChatGPT, Claude, or similar). Notice the generic answer. Then rewrite the prompt with all the specific context from the "structure" section above. Ask the same AI tool the detailed version. The difference will be striking—and you'll understand why prompt engineering matters.
Peri can explain this concept, give practical examples, help you decide whether it applies to your situation, or recommend a journey if appropriate.
Explore related journeys or tell Peri what you're working through.