Mobile · capture01
Fast enough for the moment.
Capture what you can while it’s fresh.
What did you have?
08:12Greek yogurt + oats421 kcal
13:24Chicken rice bowl684 kcal
16:48Protein shake183 kcal
Scan
Photo
Voice
Evidence-gated adaptive nutrition
Your plan should learn from you.
Track what you eat and how your body changes. Lenuri is designed to adapt nutrition from your real response, and to hold when the evidence isn’t good enough. Strength training is designed to follow the same evidence-first model and is in development.
Private beta · 18+ · No spam · Unsubscribe anytime
Example decision trace
Current plan
2,310kcal
Window 35 / 35
Observed
2,243 kcal avg
−0.43 kg / week
Confidence
Strong
84%
Decision
ADAPT
2,310 → 2,280
Explanation
The trend is stable and the window contains enough reliable logging to support a small adjustment.
How it works
A target is a starting hypothesis.
Formulas can give you a useful starting point. Your own response decides what happens next.
- 01Estimate
Start with a useful baseline from the information available about you.
- 02Observe
Collect nutrition, weight and relevant context without pretending every input is perfect.
- 03Evaluate
Read the trend over time and grade whether the evidence is reliable enough to act on.
- 04Adapt or hold
Change the plan when the evidence supports it. Otherwise, leave it alone.
Real-world logging
Log now. Repair later.
Real life isn’t perfectly logged in real time. Lenuri is being built around that reality.
Desktop · recover02
Fix what you missed when you have time.
Better data without pretending perfect logging is realistic.
MONComplete dayverified
TUELunch missingrepair
WEDDinner estimatedreview
THUComplete dayverified
FRIBackfilled +2 dayslow confidence
3 days need attention.Open queue →
Not every signal deserves a reaction
Sometimes the smartest adjustment is no adjustment.
Observed
3 incomplete days
Trend quality is ambiguous.
Confidence
58%
Decision
HOLD
2,310 kcal stays unchanged.
Explanation
There isn’t enough reliable data to support an adjustment yet. Keep logging normally; a later completed window can support a decision when the evidence is strong enough.
Designed for imperfect data
Missing data isn’t a moral failure.
You forgot lunch. Ate somewhere without nutrition information. Reconstructed yesterday from memory. Changed a portion later.
Lenuri is being designed to preserve those differences instead of treating every number as equally reliable.
08:12Breakfastweighedhigh
13:24Lunchphoto estimatemedium
16:48Snackbarcodehigh
19:40Dinnerbackfilled +2dlow
Illustrative confidence78%
The design goal is to keep uncertainty visible, so weak data is not silently promoted to certainty.
One system
Nutrition and training belong in the same feedback loop.
Nutrition is the first closed loop. Strength is in development, designed to follow the same evidence-first structure.
Nutrition
Intake2,243 kcal
Weight trend−0.43 kg/wk
Coverage31 / 35
Recoverystable
Observed
Your response
One evidence model across both systems.
Training
Sets3 × 10
Load60 kg
RIR2
Recoverynormal
Strength system in development
Evidence gate Adapt or hold
Next nutrition plan
2,280 kcal
Planned training output
maintain / progress
Principles
Less certainty. Better decisions.
01
Evidence over urgency
Lenuri shouldn’t change something merely because it can.
02
Real life over perfect compliance
Corrections, missing meals and uncertain estimates are part of the system, not failures outside it.
03
Explain the decision
A recommendation should come with a reason you can understand.
04
Safety can say no
Some situations should pause or remove a recommendation instead of producing a more aggressive one.
Privacy by product design
Your data is the product input. It isn’t the product.
No sale of health data.
No behavioral advertising built from health data.
No advertising pixels inside private health experiences.
These are product commitments. The waitlist form asks only for an email address; normal hosting and security logs may contain limited technical request data.
Private beta
Build a plan that learns when to change.
Lenuri is currently in private development. Join the early-access list to follow the build and get invited when testing opens.