ARC ENGINE
Your body changes.
The plan should too.
ARC Engine connects what happened, what you did and what happened next — then uses that context to improve the next decision.
ARC = Adaptive Response Core · the adaptive intelligence layer behind ESIVA.
WHAT ARC DOES
It doesn’t just
read signals.
Sleep affects recovery. Recovery affects training. Training affects appetite. Meals affect hunger. Tomorrow starts with all of it.
ARC follows consequences.
01
Observe
What happened?
02
Understand
What changed around it?
03
Guide
What makes sense now?
04
Watch
What happened afterwards?
05
Adapt
Should the next decision change?
RESEARCH ORIGIN
Before ARC learned users,
we had to learn the problem.
The first metabolic research pilot followed two real participants through ordinary eating, activity, sleep, training and recovery.
The goal wasn’t to prove universal rules.
It was to discover which questions were worth asking.
ARC ENGINE
MEALS
SLEEP
RECOVERY
HUNGER
TRAINING
WEIGHT
Continuous glucose monitoring was used during the pilot as an additional observation layer — not as the foundation of the future product. ARC Engine was designed around a non-CGM core so that the system could work from signals available to ordinary users.
WHAT THE PILOT CHANGED
A single metric
rarely told the story.
Similar-looking events could lead to different outcomes depending on the rest of the day.
Not universal rules. Patterns worth testing.
Hunger was bigger than glucose.
Meal adequacy, timing and context could matter as much as an isolated glucose response.
Training changed what “enough” meant.
Food and recovery could not be interpreted without knowing what training had happened — or what was still ahead.
Sleep changed the next day.
Recovery context could alter appetite, energy and the usefulness of the original training plan.
Timing mattered.
A sequence of small choices could sometimes explain the outcome better than one isolated meal or workout.
Context had to stay connected.
Meals, training, hunger, sleep and recovery made more sense as a chain than as independent dashboards.
FROM RESEARCH TO SYSTEM
The research didn’t
become a rulebook.
It became structured reference knowledge for ARC Engine: hypotheses the system could start from, then test against the individual user.
21
seed patterns
21
recommendation hypotheses
63
alert policies
9
check-in families
Seed knowledge ≠ universal truth.
Every user has to build their own evidence.
Every user has to build their own evidence.
THE MISSING FINAL CHAPTER
The pilot ended.
The learning didn’t.
Tester 01 and Tester 02 became the first reference basis for ARC Engine’s metabolic knowledge layer.
Their results were not copied into a prescription for everyone.
Does this user get hungry sooner after this kind of meal?
Does training repeatedly change their meal timing?
Does poor sleep change appetite or training quality for this person?
Did the recommendation actually improve the next outcome?
Is a pattern repeating — or was it simply coincidence?
THE PRINCIPLE
Research gives ARC
somewhere to start.
Real users decide where it goes next.
ARC Engine starts with observed patterns, watches what happens in the individual, and earns personalization over time.
A missing response is not automatically a contradiction. One unusual day is not automatically a pattern. Personalization should come from repeated evidence.
Built from real days.
Designed for yours.
ARC Engine powers the adaptive decision loop behind ESIVA.
See ESIVA in action ↗