rec.001 src: wearable + lab panel you own model: frontier AI fee: no new subscription
Early, invite-only beta

What you'd pay Whoop, Oura, and Function for — on the Claude plan you already have.

Readiness, recovery, training, and biomarker coaching — read from the wearable and lab panel you already own no sensor, no blood draw, no clinician of ours, reasoned by a frontier AI model, on the Claude plan you already pay for. Your records stay in files you own. Every recommendation carries its study and its strength of evidence. Nothing sold.

VO₂max → mortalityTier 1 ApoB targetpeer-reviewed PhenoAgeTier 2 cold plunge, NMNflagged Tier 3–4
Frontier-AI coaching No new subscription Your data, your files Cites its evidence

Coaching, not medical care — no longevity or anti-aging promises; the name is the long view, not a claim on your lifespan. Ilfirin runs on your own computer, reasoned by Claude (a frontier AI model) through the Claude plan you already pay for — so "no new subscription" means no extra fitness-app fee, not "free": you still need a Claude plan and an always-on machine. Privacy by architecture — no Ilfirin cloud, account, or database. Your records stay in files you own; during a chat, the readings Claude needs pass through Anthropic. Reads today: Fitbit · Pixel · Withings · Garmin (community); Whoop and Oura addable; Apple Watch and CGM not yet.

01 How it works

It reads the instruments. A frontier model reads the evidence.

Most of the passive signal — heart rate, sleep, steps, recovery — is read for you, with no form to fill. The rest is a short chat. Inputs in, a cited reading out.

Inputsin
Wearables — Fitbit · Pixel · Withings · Garmin auto
Meals — snap a photo or say it ~30s
Bloodwork — lab PDFs you upload annual
Check-ins — once a day ~30s
ilfirin
on your machine · in a Claude chat
A five-domain, evidence-based framework — Movement, Recovery, Fuel, Biology, Environment. Every recommendation carries its source and its evidence tier. Nothing sold.
no composite scorecomponents showncited
Outputsout
An adaptive weekly plan wkly
A daily readiness call daily
The five-domain dashboard live
Biomarker & PhenoAge trends per lab
A weekly review wkly

The full read path, step by step

02 · Why it's different

Three things the paid apps won't put in writing.

Stated plainly, because each one is checkable.

01 · evidence

It shows the paper — and how strong it is.

Every recommendation ships with the study behind it and an honest tier, from Tier 1 (large cohorts, well settled) to Tier 4 (early, unproven). "Research-backed" is an adjective. Ilfirin reports which research, and how far it actually reaches.

Tier 1 Tier 4
02 · incentive

It sells nothing it recommends.

No supplements, no prescriptions, no marketplace. That is exactly why it can flag NMN, cold plunges, and time-restricted eating as the Tier 3–4 bets the evidence says they are — rather than inventory to move.

NMN · Tier 3–4
03 · architecture

Your records live in files you own.

It runs on your own computer; your data is plain files on your machine, with no Ilfirin account and no company database. Privacy by architecture, not by promise — during a chat, the readings Claude needs pass through Anthropic, and we say so.

no account no database
03 · The data layer

It reads the wearable and meters you already own.

Your watch, scale, and cuff land in one view automatically — you're not transcribing numbers by hand. The device list is stated exactly, including what isn't supported yet.

Fitbit Pixel Withings — scale + BP Garmin (community) Whoop — addable Oura — addable Apple Watch — not yet CGM — not yet
Reads today

Fitbit · Pixel · Withings · Garmin (community). Most of the passive longevity signal is captured with zero ongoing input. The remainder — food, bloodwork, and how you actually feel — is a normal chat, in seconds. If you're filling out a form, we failed.

04 · Meal logging · photo-first

The one input you add by hand — now it's just a photo.

Your watch and scale report themselves; food is the part you log. So we made it the lowest-friction input there is — snap the plate, answer one question, done — and kept it honest about what a photo can and can't read.

01 · names what it sees

Identifying food is what photos do well.

Ilfirin lists what's on the plate for you to check — salmon, rice, broccoli — before any math. Recognizing what a food is, is the thing image models do reliably; across published tests, food identification is reliable — accuracy in the 90s. That's most of what the diet-quality score needs.

02 · asks the one thing it can't

Portion is the weak spot, so it asks.

A photo can't size a serving or see the oil in the pan. Rather than guess, Ilfirin asks the single highest-value question — usually just the portion — and one confirmed answer collapses most of the error. The question is the honesty signal.

03 · a range, not a fake number

Shows a range until you confirm.

Until you confirm the portion, Ilfirin shows protein as a range, not a single figure dressed up as truth — confirm it and the range closes to a number. Micronutrients come from the matched food record, never the pixels; what it can't know reads "unknown," never a silent zero.

Not "AI counts your calories from a picture." That claim is everywhere, and on real mixed meals it doesn't hold — independent testing of photo-calorie apps lands well short of it. Ilfirin's design is the one the research actually supports: photo for the identity, one question for the portion, a confirm step before anything is written, and a range until then. A meal photo is chat content — in standard mode it passes through Anthropic like anything else you type.

See a real photo log, start to finish

05 · The framework · five domains

Five domains. Every recommendation cites its study and its tier.

No single composite number — components, shown. These are the same five cards you read on the dashboard.

Movement how fit and strong you are

VO₂max is the strongest single reading we track — it predicts all-cause mortality more strongly than smoking or high blood pressure, with no observed ceiling (Mandsager 2018, JAMA Netw Open). Grip strength is a separate marker. The resistance-training sweet spot is a modest 30–60 min/week.

Tier 1 e.g. VO₂max trend
Recovery how well you sleep & bounce back

~7–8 h, and regularity may matter as much as total (Windred 2024). Heart-rate recovery and HRV are honest but noisy autonomic signals — read as trends, not verdicts. Loud snoring or witnessed apneas → test for sleep apnea; untreated it shortens lives, and it's treatable.

Tier 1 e.g. sleep-regularity index
Fuel what you eat and drink

The Mediterranean pattern is the most RCT-supported way to eat for the long view (PREDIMED). Enough fiber and protein; less ultra-processed food. On alcohol, the health-optimal dose is zero — we report the evidence and don't moralize about it.

Tier 1 e.g. protein adherence
Biology what your bloodwork says

ApoB is the headline marker — causal for atherosclerosis, so lower reads better. Plus Lp(a), inflammation (hsCRP), glycemic control (HbA1c), and blood pressure. PhenoAge is computed as a biological-age trend — never a precise number, and labelled Tier 2. No $300 epigenetic "aging clocks."

ApoB · Tier 1 PhenoAge · Tier 2
Environment the conditions you live in

Weak social connection carries a mortality signal comparable to smoking (Holt-Lunstad 2015). Plus air quality (no safe threshold; modern targets ~5–9 µg/m³ PM2.5), purpose, cognitive engagement, and a calendar of proven age-appropriate screenings.

Tier 1 e.g. screening calendar
read the source

Every citation, in full.

The complete framework — five domains, each target with its peer-reviewed source and its honest tier Tier 1 settled · Tier 4 early, unproven.

06 · Privacy · by architecture

It runs on your computer. Your records stay there.

Like a calibrated meter a trainer brings to your house — it reads your own notebooks on site and takes no copies home. The opposite of mailing your records to a company you've never met.

mode · standard

Best AI quality

Records stay on your machine; during a chat the readings Claude needs pass through Anthropic, so it can coach with full detail. The default.

mode · aggregated

Summaries only

Raw values stay local. Claude reads "within range" or "trending up" — not your exact ApoB or weight. Some granularity traded for a narrower path.

mode · local-only

Nothing leaves

A model runs entirely on your machine. AI quality drops; the privacy is absolute. Your call, stated as the real trade-off it is.

Honest by default. No Ilfirin cloud, account, or database — there is nothing for us to sell or leak. Your records live in files you own. On standard mode, the readings Claude needs pass through Anthropic during a chat; we link Anthropic's privacy policy and don't pretend otherwise. Three modes, field-level redaction, and a local audit log of every data access — readable any time.

See exactly what stays local and what doesn't

07 · See it in action

Faithful, anonymized readings — not a demo dressed as a product.

Walk the real instrument on sample data: the local five-domain dashboard, the single actionable call, and meal logging that cites its sources.

view · dashboard

The full dashboard

Five domains, one view — components shown, never a single composite score. The headline is the one signal worth acting on this week.

view · insight

Insights you act on

The one thing to do next — with the study and the tier carried alongside it, so you can see how far the recommendation reaches.

view · food log

Photo-first meal logging

Snap the plate — it names the food, asks the one thing a photo can't size, and shows a range until you confirm. Or type it; same machinery, sources cited, guesses flagged.

See Ilfirin in action

Request access

Want it set up for you?

Tell us your devices and your setup. We provision your own instance and send a personalized installer — no terminal required. Access is limited while we keep the beta small.

An early, invite-based beta — not a finished, polished product, and not pretending to be. Coaching, not medical care. Immortality is the name, not the offer — Ilfirin coaches toward the long view, with no claim on your lifespan.