Technical evidenceRaw pressure times and Process S
Pressure plan—Raw pressure and age-band values.
Forecast median—
Named model view
Process S—Pressure plus recent timing.
Prediction league
Every forecast is locked in before the sleep, then scored automatically once it is logged. Each sleep counts once, on the forecast that was standing when it happened.
Loading prediction scores…
Recent bets
Loading predictions…
Settings
Account
Signed in as
Get a code to type there. No email needed.
Sign-ins
Not set
Sign in with Apple on this account too.
Sign in with Google on this account too.
Download everything logged in every family you're part of.
Removes your account everywhere. If you're the only one in a family, that family and its history go too.
Family
Children
Prediction league
Keep one to three reproducible algorithms active. Each new bet records all of them at once, on identical information.
Pressure physics, personalized by recent timing.
Median clock time for this nap position or bedtime.
Median awake interval from the latest wake.
AI second opinions
NaniNani · Process-S engine fitted on 671 days of real sleep logs.
A pressure gauge, never a deadline. Not medical advice.
Development notebook →
NaniNani
See when sleep is coming.
Less to keep track of, not more: one calm glance shows when sleep is likely.
No account needed to try.
THE SHORE
While your child is awake, the water rises.
That’s sleep pressure: the body’s drive to sleep, building hour by hour. Sleep drains it again.
THE SHOREsleep comes easilythen a next wave
Near the shore, sleep comes easily.
Too early, and there’s little to release. Too late, and settling takes longer. If a wave passes, another one comes.
A promise, before we start.
NaniNani is here to take things off your mind.
Times are windows, never deadlines.
No streaks, no scores, no nagging notifications. Nothing here grades you.
Your child’s cues come first. The screen comes second.
No account needed. To keep logging after 30 days, save it to an account; without one, data unused for 60 days is deleted.
When did your child wake up today?
The tide starts here. Roughly is fine.
Today
07:00
Join a family
Enter the 6-character code you were given.
Family
Members
Activity
Change access
What should your family call you?
They see it next to what you log, and with any invite you send. Only people in your family see it.
Invite someone
Choose which children they can see. Good for 7 days.
Anyone with the link or code can join -- share it only with who you mean to invite.
Delete your account
This can't be undone. Deletes your account everywhere, along with any family where you're the only member — its children and events go too. Families you share with others keep their history; if you're the only one with full access to a child there, give someone else full access first.
If you signed in with Apple, you can also remove NaniNani from your Apple ID: Settings → your name → Sign-In & Security → Sign in with Apple.
How timing works
A guide, never a deadline
NaniNani shows two times for the next sleep. One is where the biology points. The other is the median of the available forecasts. The distance between them is the part worth reading.
The two times, and what each one answers
Prescribe answers “where should we aim?” Predict answers “what will probably happen?” Prescribe is a pressure-based target; Predict takes the robust middle of the forecasts that can answer today.
PrescribeWhere to steer
—Pressure and the body clock, held inside this age’s evidence band.
PredictWhat is likely
—The middle of the available Process-S, clock-anchor, wake-window, and Napper forecasts.
Live guidance appears while the child is awake and enough data is available.
What the percentage on the gauge means
The number is accumulated sleep pressure measured against the model’s average sleep-onset threshold. Near 100%, pressure has reached the level the model treats as the usual point where sleep comes more easily. It is not a signal that sleep has to happen now.
Pressure climbs while awake, fast at first and then more slowly, and drains while asleep. The real threshold moves too: the body clock raises it through the late afternoon and lowers it overnight. The gauge uses the flat average so the number only ever rises while awake, while Prescribe uses the moving version. That is why a rising gauge and a shifting target can feel slightly out of step.
The first months play by different rules
A newborn’s sleep runs on cycles of roughly an hour, driven by feeds, scattered around the clock. That is the design, not a problem. The 24-hour body clock switches on later: a melatonin rhythm first appears around 9–12 weeks and steadies by 13–15, and day-night structure only becomes definite after the third month. Until then, days that look nothing like each other are biology doing exactly what it should.
NaniNani follows the same timeline. Under two months it never proposes steering — there is no clock to steer by — and while the clock is switching on it leans on sleep pressure and stays loose about clock times. Two things worth knowing besides: milestones count from the due date, so a baby born early is “younger” than the calendar says; and colicky babies (about one in five) run this whole timeline later, with drowsy cues that are hard to read. None of it is caused by anything you’re doing.
The forecasts that get scored
Each of these forecasts the actual onset, and each one is written down before the sleep happens. That is what makes the comparison fair: no algorithm gets to see the answer first.
Process S + habits
Starts at the Prescribe crossing, then adds the recent gap between that target and when sleep actually started, kept separate for each nap position and for bedtime. The most personal of the three, which is also its weakness: it will happily learn a late drift you were hoping to correct.
Clock anchor
Takes the middle clock time of comparable recent naps or bedtimes and repeats it. Steady when the days are regular, stubborn when they are not. An hour-early wake-up moves it by nothing at all.
Wake window
Takes the middle awake stretch before this sleep and counts it out from the morning wake or the end of the last nap. It follows a shifted day well. It knows nothing about the body clock, or about how much the last nap actually drained.
Why Prescribe stays out of the league
Prescribe does not compete in the league, and it would score badly if it did. It was never describing today. It describes where the biology points, and a week where sleep moves toward that target is a week where the guidance did its job, even though every single bet would have looked wrong on the day.
Scoring starts at 100 for an exact hit and is deliberately lopsided: a forecast that ran early loses 1.5 points per minute, one that ran late loses 2.5, because a late forecast steers the parent toward overtiredness and an early one merely costs some waiting. Naps and bedtimes are kept apart, and coverage records how often an algorithm had enough history to answer at all.
Using the two times day to day
Use the hero’s start-helping countdown for today’s next move: it starts from the likely onset and backs off by this child’s learned help-to-sleep interval, or the current family fallback while there are fewer than three answers.
Read Prescribe as a pressure-based target to steer toward, and Predict as today’s likely onset. Later than Prescribe means the available forecasts collectively point later; earlier means they collectively point earlier.
When the separate Process-S habit drift holds for three days or more, the steer line proposes one small move — at most 20 minutes earlier per night, 15 later — and asks you to hold it for three nights. One odd day never triggers it.
Log onsets and wakes as they really happened, messy ones included. When you can, answer the optional question about when you started helping them sleep; after three answers, the app uses that child’s own typical interval for naps or bedtime.
The reasoning behind all of this — the research, the principles, the honest limits — lives in the development notebook.
Good timing makes settling easier and protects a decent sleep opportunity. It does not decide how well a child sleeps — illness, teeth, the room, and the child’s own needs all outrank it, and so do their cues in the moment. NaniNani is not medical advice.
Algorithm
One algorithm at three depths: the idea, how to read it on a normal day, and what the code actually computes.
Preview app states (dev)
☁️Nap
Helping sleep since
Track time spent helping to sleep
Start
End
Still sleeping
About how long did it take?
From when you started helping them fall asleep until they were asleep. An estimate is enough.
Start date
Almost always today — you only need this for sleep that crossed midnight.
Children
Edit profile
NaniNani uses corrected age from the due date until 2 years.
Add a child
NaniNani uses corrected age from the due date until 2 years.
Sign in another device
Type these on the other device within 10 minutes. It works once.
How this was worked out
Today, so far
Keep what you've logged — save it to an account.
NaniNani
Sign in to your household
Signing in with Apple or Google switches to that account. This device's own data stays with the anonymous account it's in now, and isn't merged automatically.
