CONCEPT PROJECT  ·  iOS MOBILE  ·  2026

Pulsehealth AI that reads between the lines

A self-initiated design exploration into agentic AI for personal health: an app that notices what your data means before you think to ask, and always shows how it knows.

Role
Solo designer
Platform
iOS mobile
Scope
4 high-fidelity screens
Methods
Research, scenarios, prototyping
Live prototype · scroll inside the phone
01
Context

Health apps track everything.
They understand almost nothing.

Most health apps are dashboards with a coat of paint. They surface numbers like resting heart rate, sleep duration and step count, and leave you to figure out what any of it means for how you actually feel. The apps that try to go further do it through generic push notifications and one-size-fits-all coaching that knows nothing specific about you.

Pulse is the other version of that. It's a fictional health companion that reads your data over time, notices patterns you'd never catch yourself, and explains them in plain language with its reasoning visible. The product is speculative. The design problem it explores is not.

“

I have all this data and I still don't know why I feel the way I feel. The app just shows me more graphs.

Sentiment shared across conversations with health tracker users
02
Research foundation

What people actually want
from health technology

I grounded the project in informal conversations with people who use health trackers daily, plus secondary research on AI adoption in consumer wellness. I wasn't asking about features. I was asking about the moments that mattered and the moments that frustrated. Four patterns shaped the design.

1
Data without context creates anxiety, not insight
People described checking their HRV or sleep score and feeling worse for it. The numbers weren't bad; they just didn't know what to do with them. Raw data without interpretation is noise.
2
People want patterns surfaced, not dashboards to explore
Nobody opens a health app wanting to do analysis. They want to be told something true and specific, especially something they hadn't noticed. The value is in the signal, not the data.
3
Generic recommendations feel like spam
"Time to move!" gets ignored because it applies to everyone. Specificity, meaning your pattern, your time window and your data, is what makes a recommendation worth acting on.
4
Trust is fragile, and one wrong call breaks it
In health, AI that oversteps loses credibility instantly. Users are very aware of the difference between a correlation and a diagnosis, and they expect the app to be too.
03
User persona

Who Pulse is designed for

A composite built from those conversations and secondary sources.

Portrait of Jordan M., the Pulse persona
Jordan M.
Product designer · 32 · Chicago
Apple Watch + Oura · logs food inconsistently

"I know my data is telling me something. I just can't figure out what."

AI attitude
SkepticEarly adopter
Goals
  • Understand why energy fluctuates
  • Build habits, not streaks
  • Get specific guidance
  • Feel like data works for them
Frustrations
  • Opens apps, feels overwhelmed
  • AI advice feels copy-pasted
  • Doesn't know which metrics matter
  • Notifications feel like interruptions
Context
  • Checks data morning and night
  • Has 4 health apps, uses 1
  • Prefers reading to dashboards
  • Skeptical of claims, receptive to data
Mental model
  • Thinks in systems, not scores
  • Values honesty about uncertainty
  • Wants to stay in control
  • Would share data if trust was earned
SPECEvery insight references Jordan's own data, time window and source, never a population average
TRUSTCorrelation is labeled as correlation. Confidence qualifiers are a designed UI element, not fine print
CTRLPulse surfaces and suggests. Jordan decides. The AI never starts an action on its own
PRIVWhat Pulse is reading is always visible, as a strip in the AI chat rather than a buried permissions screen
04
Design principles

Four principles that
shaped every decision

I committed to these before touching a screen. Every design decision in the prototype traces back to one of them.

01
Specific over general, always
"Your HRV drops 18% between 1:45 and 3:15pm" is far more useful than "afternoon slumps are common." Every insight references your data, your window and your pattern.
02
Show the reasoning, not just the result
Every insight carries a one-line trace: what data it used, what it compared, and how confident it is. Transparency is the product, not a disclaimer.
03
Pulse notices. You decide.
Pulse never sets reminders, starts experiments or changes anything until the user says so. The agentic moment is always started by the person.
04
Honest about the limits of data
Confidence is visible. Pulse never diagnoses, prescribes or claims certainty it doesn't have. That's responsible design, not legal cover.
05
The work

Four screens.
One thread running through them.

Each screen answers a different design question, and the same insight runs through all four: Pulse notices that the user's 2pm energy dip is tied to their morning coffee. Following one finding from first mention to action shows how trust is built step by step. These are the live prototypes, so you can scroll inside each phone.

Sage
Anything Pulse says or does
Gold
Sleep
Lavender
Recovery and HRV
Sky
Stress and caffeine
Rose
Dips and outliers to look at
Fraunces · DM Sans · warm off-white
Good morning,
you slept well.
A soft editorial serif for the moments Pulse speaks to you, and a clean sans for data. Calm and warm on purpose, so it reads like a journal about you rather than a fitness tracker.
Scroll inside
SCREEN 01 · HOME

A morning briefing, not a metrics dump

How does AI earn attention without demanding it?

The home screen leads with a sentence, then one orienting score, then a single proactive insight. Everything else is there if you want it, but the story comes first.

  1. 1The greeting is the headline. "Good morning, you slept well" tells you how you're doing before you see a single number.
  2. 2The score explains itself. 76 comes with its reason: best in two weeks, because sleep and HRV both improved overnight.
  3. 3"Pulse noticed" card. One insight, with its reasoning right on the card: compared HRV, caffeine log and steps over 14 days. "Correlation is strong, not causal."
  4. 4Evidence on every card. Insights show how much data they rest on (5 nights, 9 sessions), and the log tags entries "Auto-logged by Pulse" so you can tell what you entered from what the AI did.
proactive AImorning brief
Scroll inside
SCREEN 02 · INSIGHT DETAIL

The transparency layer, made visual

What does it take to trust an AI's conclusion about your own body?

Tapping the caffeine insight opens the full story: the pattern, the evidence behind it, and what you could do about it.

  1. 1The chart makes the claim visible. The 1:45 to 3:15pm dip window is highlighted in rose so you see the pattern in a couple of seconds.
  2. 2"How Pulse found this." A four-step trace where every step names its source, like HealthKit HRV or the caffeine log.
  3. 3It tests the counter-case. Step 3 checks the days that break the pattern: on noon-coffee days the dip was 40% shallower. Looking for disconfirming evidence is what makes the conclusion believable.
  4. 4Offers, not actions. A 1:30pm walk reminder and a 5-day noon coffee test. Neither starts until the user picks one.
transparencydata viz
Scroll inside
SCREEN 03 · TRENDS

Patterns over time, already interpreted

How do you show change over a week without asking people to do analysis?

Trends is the bigger picture. Pulse writes the summary, and every chart below comes with a one-line takeaway. The dark banner separates Pulse's narrative from the raw data.

  1. 1The summary names the cause. Recovery is up because 6 of 7 nights fell within the same 30-minute bedtime window.
  2. 2One outlier, flagged everywhere. Tuesday's late night shows up in rose on the sleep chart, and the slower recovery shows up on Wednesday's HRV, with the cost spelled out: 40 fewer minutes of deep sleep.
  3. 3Every chart has a takeaway. Each metric card ends with a single line from Pulse, like workout days showing 18% lower stress.
  4. 4A month at a glance. The 28-day heatmap makes the low mid-week days stand out without reading a single number, and today's date ends the grid.
patternsheatmap
Scroll inside
SCREEN 04 · PULSE AI

A conversation that ends in action, on the user's word

What does agentic AI look like when it acts on health data?

The user asks why they're tired at 2pm. Pulse answers with the same evidence as the insight, handles a follow-up worry with the user's own data, and only acts once they say go.

  1. 1"Reading:" strip. Sleep, HRV, Activity, Journal. You always know what Pulse is looking at in this conversation.
  2. 2Evidence inside the answer. An inline data card, plus the line "This is a pattern, not a diagnosis."
  3. 3Personal, not generic. Asked whether later coffee hurts sleep, Pulse checks the user's own bedtimes (11:04 vs 11:02pm) instead of quoting general advice.
  4. 4Acts only after "Let's do it." Then it confirms exactly what it will track, the dates, the reminders, and when results arrive on the 28th.
agentic AIconversational
1

The reasoning trace, inside every card instead of behind a tap

Every Pulse insight shows a one-line reasoning trace inline rather than behind a "how did we get this?" button. In health, trust isn't built by the insight alone. It's built by the evidence trail.

A user who sees "Compared HRV readings, caffeine log, and step data across 14 days" can evaluate that claim. A user who sees only the conclusion has to take it on faith.

Design insight: Users don't need to read the trace every time. They need to know it's there every time.
2

Correlation vs. causation, named explicitly in the UI

Every cross-referenced insight includes a visible qualifier: "Correlation is strong, not causal." It's a designed element, not fine print. Health AI that overstates its certainty even once can lose the user for good.

Showing the limits of what Pulse knows is a feature. It's what makes every other claim more believable.

3

The 5-day experiment, structured autonomy

When Pulse finds a pattern you can do something about, it proposes a structured experiment: one specific change, a defined timeframe, and a promise to track and report results. It's the most deliberate agentic choice in the project.

It gives the AI real work to do on the user's behalf while keeping the person in control of whether it happens. Pulse designs the experiment. The user decides to run it.

4
Screens, one insight followed from first notice to action
0
Actions Pulse takes without the user saying yes
1
Traceable source behind every insight. No outputs without inputs.
Design outcomes
Design goalHow it was addressedWhere to see it
Insights feel personalEvery card references the user's own time window, duration and data sourceAll 4 screens
AI reasoning is visibleA reasoning line shows inline on every Pulse surface, no tapping requiredScreens 1, 2, 4
User stays in controlPulse offers; nothing starts until the user chooses itScreens 2, 4
Data anxiety reducedHome leads with a plain-language read, not a wall of metricsScreen 1
Change over time is readablePulse-written summary plus a one-line takeaway on every chartScreen 3
Trust built step by stepConfidence qualifiers and "pattern, not a diagnosis" are permanent UI fixturesScreens 1, 2, 4
Reflection · what I'd do next
Design the experiment results screen
The 5-day experiment is the most interesting agentic idea here, but I designed the setup, not the payoff. What does the results screen look like when the outcome is inconclusive? That's the hardest screen to get right and probably the most important.
Give privacy its own screen
Pulse reads sleep, HRV, activity, caffeine and mood. The "Reading:" strip shows what it's using, but a real product needs a full privacy surface: what Pulse has read, what it has inferred, and how to delete any of it.
Test with heavy health trackers
The persona came from real conversations, but the prototype hasn't been tested with people who track obsessively. They have strong opinions about what counts as a meaningful insight and where an AI crosses a line, and their edge cases would make the design sharper.