Circa Get early access

The science

We tell you why, every time, with citations.

Circa runs a real, unit-tested circadian engine. Every recommendation has a reason and a source. Every claim is graded by evidence strength. The thing we will not do is overclaim.

Core: settled science Supportive: well-evidenced Comfort: environment

Built on real circadian science

  • 648

    Plans tested

    108 routes × 6 profiles, 100% correct

  • 82

    Unit tests

    JS engine, all passing

  • 8 levers

    In every plan

    Light, sleep, caffeine, melatonin, exercise, food, hydration, transit

  • 0

    Pseudoscience tolerated

    Evidence-graded recommendations

The eight levers

Every lever the science supports. Graded honestly.

Settled science is tagged Core. Well-evidenced but more modest effects are tagged Supportive. Environment and comfort moves are tagged Comfort. No overclaiming, anywhere.

  • Core

    Light, timed in both directions

    Bright light at the right hour of your biological day advances or delays your clock by an hour or more. Wrong time, it does the opposite. Direction matters; Circa handles it.

  • Core

    Sleep, including on the plane

    A two-process model decides when sleep is biologically available. Circa knows when to sleep aloft to protect your destination night, and when staying awake is the right move.

  • Core

    Caffeine, cut at the right hour

    Caffeine has a five-to-six hour half-life. Mistimed, it sabotages the sleep window you are trying to protect. Circa schedules your last cup against the right reference.

  • Core

    Low-dose melatonin, phase-anchored

    0.3 to 1 mg taken at the right destination hour shifts the clock. Mistimed, it does nothing or backfires. Melatonin is a chronobiotic, not a sleeping pill.

  • Supportive

    Exercise, the underused zeitgeber

    Movement at the right time of the destination day reinforces the rest of the protocol. Placed wrong, it confuses the signal. Circa places it relative to your real schedule.

  • Supportive

    Meals, on destination time

    Peripheral oscillators in the gut and liver respond to meal timing. Circa schedules destination-zone meals in flight, not after arrival.

  • Supportive

    Hydration and electrolytes

    A long-haul flight can cost two to three litres of fluid. The dehydration alone produces fatigue that looks like jet lag and stacks on top of it.

  • Comfort

    Transit, alcohol, illness-prevention

    When to mask, when to cut alcohol, what to ask the cabin crew, when a lounge shower beats a hotel nap. Every plan covers the real-world stuff most apps ignore.

The model

Two processes, two phase-response curves, one usable plan.

Process S and Process C

The two-process model, originally formalised by the Swiss sleep researcher Alexander Borbely, describes sleep as the interaction of a homeostatic process (the longer you have been awake, the more pressure to sleep) and a circadian process (alertness oscillates on a near-24-hour rhythm regardless of how long you have been awake). Together they determine when sleep is biologically available and when it is not. This is why a Circa plan can recommend sleep on the plane during your destination’s biological night, rather than telling you to stay awake the entire flight.

The phase-response curves

The phase-response curve (PRC) for light describes how the body clock shifts in response to light at different times of the biological day. Light after your core body temperature minimum (CBTmin) advances the clock; light before CBTmin delays it. Melatonin is the inverse: evening doses advance the clock when timed correctly. The PRC framework is settled science and is the same framework Timeshifter and other respected jet-lag tools use. We just apply it more thoroughly.

Daily shift magnitude

The unaided circadian system shifts roughly 0.5 to 1 hour per day. With well-timed light, melatonin, sleep, and exercise, the daily shift can be roughly doubled. The eastward direction is harder than westward, because advancing the clock fights the natural drift. Magnitude depends on direction, prior light exposure, age, chronotype, and where CBTmin sits at the start of the trip. None of these numbers move on their own; the protocol moves them.

What we explicitly do not promise

  • We do not claim to cure jet lag. Jet lag is not a disease.
  • We do not promise zero adjustment. Biology imposes a maximum daily shift no protocol exceeds.
  • We do not promise the same outcome for everyone. Personalisation is the USP precisely because individuals differ.
  • We do not make medical claims for supplements. Copy stays structure/function with disclaimers.
  • We do not claim the supplements are required. They are recommended where the engine calls for them and tagged by evidence strength.

Engineering

A real engine. Unit-tested. The source of truth.

82

JS unit tests

Every code path in the circadian engine is exercised by a deterministic test, runnable on Node with zero dependencies. CI blocks any merge that breaks a test.

648

Plans tested in an eval harness

108 routes times 6 traveller profiles. We grade every output for correctness against the published PRC literature. CI gates at 99% correctness; current is 100%.

2

Identical engines, iOS and Node

The Swift port mirrors the JS reference one-to-one. Both pass the same tests. The engine on your phone is the engine we tested.

If you want to go deeper

A few honest places to start.

Coming to the App Store

Plan your next trip with Circa.

First plan free. HK$149 a year for unlimited trips, or HK$38 per trip.

Get early access Join the waitlist