What the AI found
“The AI observed that a client's carbohydrate intake on 'stressful' days (reported HRV below 25 ms) consistently dipped by an average of 45g compared to 'calm' days, suggesting an unconscious dietary shift under pressure.”
Before
Client data across 4 apps, no clear patterns
After
AI-powered insight, ready for coaching
The same system, three states — real screens, not a screenshot
| Date | Food logging app data (daily macros) |
| Date | Oura Ring data (daily HRV) |
| Date | Client mood journal (self-reported stress) |
| ... |
Prompt
Here are 60 days of [Client Name]'s daily data, organised by date. Each row contains their daily macronutrient intake (carbs, protein, fat in grams), and their average daily Heart Rate Variability (HRV in ms) from their Oura Ring. Identify any statistically significant patterns or correlations between their HRV values (indicating physiological stress/recovery) and their macronutrient intake. Specifically, look for how macro ratios or total grams shift on days with HRV values below 25 ms versus days with HRV above 50 ms. Present any findings with concrete numbers and averages.
Here are 60 days of [Client Name]'s daily data... [pasted raw data] ...Present any findings with concrete numbers and averages.
AI
Analysis of the 60-day dataset reveals a notable pattern. On days where [Client Name]'s HRV was below 25 ms (indicating higher physiological stress), their average daily carbohydrate intake was 160g. This contrasts with days where HRV was above 50 ms (indicating better recovery), where carbohydrate intake averaged 205g. This represents an average reduction of 45g of carbohydrates on higher-stress days.160g
Avg. Carbs (High Stress Days)
205g
Avg. Carbs (Low Stress Days)
-45g Carbs
Daily Macro Shift (Stress)
One Patient, One Prompt: Uncovering a Longevity Trend
From scattered data to a concise, actionable dietary pattern analysis for longevity in under 10 minutes.
A nutritionist running a small Northern European practice
Tools used
The real tools used here — swap any for your own equivalent. Each links to how we’d set it up.
Starting state
Before anything was set up
Before implementing a structured AI review, the nutritionist gathered client data from various sources: a food tracking app, a wearable device for HRV, and weekly self-reported mood journals. This information remained in disparate silos, making it nearly impossible to spot consistent, actionable patterns without hours of manual review. Each client presented a fragmented picture, delaying effective personalised guidance.
| Date | Food logging app data (daily macros) |
| Date | Oura Ring data (daily HRV) |
| Date | Client mood journal (self-reported stress) |
| ... |
Working state
Setup, doing its job
The nutritionist consolidated 60 days of one client's data into a Google Sheet. They then used Gemini to analyse the correlation between HRV, a proxy for stress, and daily macronutrient intake. The key was a precise prompt, enabling the AI to identify subtle dietary shifts linked to physiological states, a task impractical for manual analysis.
Prompt
Here are 60 days of [Client Name]'s daily data, organised by date. Each row contains their daily macronutrient intake (carbs, protein, fat in grams), and their average daily Heart Rate Variability (HRV in ms) from their Oura Ring. Identify any statistically significant patterns or correlations between their HRV values (indicating physiological stress/recovery) and their macronutrient intake. Specifically, look for how macro ratios or total grams shift on days with HRV values below 25 ms versus days with HRV above 50 ms. Present any findings with concrete numbers and averages.
Here are 60 days of [Client Name]'s daily data... [pasted raw data] ...Present any findings with concrete numbers and averages.
AI
Analysis of the 60-day dataset reveals a notable pattern. On days where [Client Name]'s HRV was below 25 ms (indicating higher physiological stress), their average daily carbohydrate intake was 160g. This contrasts with days where HRV was above 50 ms (indicating better recovery), where carbohydrate intake averaged 205g. This represents an average reduction of 45g of carbohydrates on higher-stress days.Use case implemented
The finished system, running on its own
With the system implemented, the nutritionist now has a weekly ritual. Within ten minutes, they can feed a client's updated data into a simple Google Sheet, prompt Gemini, and receive a concise summary of any emerging patterns. This allows for proactive, evidence-based dietary adjustments, moving beyond anecdotal observations to data-driven coaching conversations.
160g
Avg. Carbs (High Stress Days)
205g
Avg. Carbs (Low Stress Days)
-45g Carbs
Daily Macro Shift (Stress)
What an outside observer would notice
reduced by 90%
Time to client insight
up 20%
Client engagement on diet
3x more
Data points analysed per session
The stack — build it yourself
Accessible, flexible for various data types, easy to export.
Reliable, passive HRV data collection for stress monitoring.
Exceptional at identifying non-obvious numerical patterns across disparate data sets.
These are the tools used in this story. Any can be swapped for an equivalent you already trust.
Go deeper
Do this yourself
See Setup
This story runs on Setup. The tools and prompts above are the real build — swap any tool for your own equivalent and follow the same steps.