Baseline at Edge Esmeralda: Residency Pilot Spec

What Edge Esmeralda gets

Ten pilot residents produce a coherent 4-week outcomes dataset: week 1 baseline, continuous home and wearable capture, week 4 follow-up, Baseline scoring against the healthspan framework. Not point-in-time observations but trajectory across the residency. Edge Esmeralda gets a structured health outcomes report from the village. Residents leave with health literacy they can act on.

Background and mission statement

Every consumer has the same core question about their health: am I healthy, and can I live healthier? Baseline is a consumer health behavioral change platform that adds a definition around chronic disease onset to answer "am I healthy," then uses that definition to drive measurable behavior change toward "can I live healthier." The mission is a healthier American consumer.

Human health has three tiers: healthspan (years lived without chronic disease), peakspan (age interval at peak functional performance), and longevity (total lifespan). Baseline starts with healthspan because "am I at risk of chronic disease onset, and can I move into a healthier range" is the strongest wedge question for most consumers. Once a user's chronic disease baseline is clean, the platform opens into peakspan-tier optimization. Longevity is a downstream tier.

Current state: 7 pilot users generating structured outcome data, 10 more onboarding this week and next. NHANES-anchored scoring across 17 axes and 65 canonical chronic diseases. Solo-plus-agents operator on a Mac Mini, local-first by default.

The gaps we're filling

• Interpretation

  Labs give numbers, not meaning. The 7-minute physical is the interpretation layer most consumers get. No infrastructure turns biomarkers into an actionable model of one's own health.

• Trajectory

  Most consumers get labs once per year at best. Longitudinal patterns are invisible because each result individually reads as "normal." Multi-year drifts into disease-risk bands go silent.

• Onset-threshold scoring

  "In range" is a false comfort. Chronic disease onset happens inside reference ranges. No consumer product today scores biomarkers against onset thresholds specifically; all report against reference ranges.

• Evidence infrastructure

  Millions are self-administering peptides, GLP-1s, supplements, hormone optimization, longevity protocols. Zero structured outcome capture at population scale. The patent-funded pharma system cannot evaluate non-patentable compounds. The substrate that could aggregate consumer-generated signal into evidence does not exist.

"In range" is a false comfort. Chronic disease onset happens inside reference ranges.

Roadmap and how Esmeralda fits

Baseline's phase sequence, north-starred to BLISS 2027 as contributor:

Edge Esmeralda is a phase-1 acceleration event and a phase-2 seed. A 4-week residency window compresses 3-6 months of recruitment and structured outcome capture into 4 weeks. Residents generate a coherent outcomes dataset the village can take home as a structured health report. That same dataset contributes to Baseline's N-tier cohort curve toward BLISS 2027. Mutual benefit: Edge Esmeralda gets evidence-grade health outcomes programming for residents; Baseline gets cohort acceleration.

Sections below specify the operational need to run the pilot: biomarker panel, cost projection, and tiered partnership asks.

The biomarker panel

The panel runs four progressive tiers, gated by chronic-disease-onset risk. Tier 0 and Tier 1 are foundational — every pilot user, always, and the scope of this proposal. Tier 2 fires only when a Tier 0 or Tier 1 signal flags onset risk, warranting deeper investigation. Tier 3 captures lifetime-stable baseline markers once at first draw. Tiers 2 and 3 are reach targets below.

Foundation panel — Tiers 0 and 1 (proposal scope)

17 measures every pilot user gets, from ongoing home capture through the default-in blood draw. This is what we are asking the partnership to cover.

Reach targets — Tiers 2 and 3

Tiers 2 and 3 are where the panel's full ambition lives. Tier 0 and Tier 1 tell us whether a user is trending toward chronic disease onset. Tiers 2 and 3 tell us why, and what specifically is driving onset risk when it fires. They add depth the pilot can carry only if partnership pricing absorbs them; otherwise the depth moves to phase 2. The cost projection below reflects Tier 2 as “if triggered” and Tier 3 as TBD, so neither is committed line-by-line at pilot scale.

Tier 2 — on-indication fires. When a Tier 0 or Tier 1 signal flags onset risk, Tier 2 investigates the mechanism behind the flag. Pre-renal function beyond creatinine, vascular inflammation and methylation signaling beyond hsCRP, thyroid autoimmunity when TSH drifts, OGTT when fasting glucose is borderline. Each fire is scoped per-user, per-event, so cost only accrues when the signal warrants it. Markers: Cystatin C, Homocysteine, B12 + folate, Iron studies (TIBC, saturation), Free T4, TPO antibodies, AAT level + phenotype, tTG-IgA (celiac serology), OGTT.

Tier 3 — one-time lifetime baseline. Captured once at first draw, these encode lifetime-stable risk context that does not move across the pilot window: genetic lipid risk, infectious disease baseline, chronic viral exposure. They are reference anchors the onset-threshold scoring pins against, not repeat measures. Markers: Lp(a), HIV serology, Hepatitis C antibody, Hepatitis B surface antigen + core.

If Invader's existing panel can absorb any of these, they move from “deferred” to “scope” for this pilot. Ideal absorption: Tier 2 fires included on-indication at no marginal cost; Tier 3 baseline captured alongside the week 1 draw. That extends the panel from onset-risk screening to onset-risk investigation in a single 4-week window.

Cost projection, per pilot user, 4-week window

At 10 pilot residents, aggregate pilot cost: ~$4,250-$7,850 fully à la carte; ~$2,450-$4,650 at partnership-tier compression. Invader-partnered aggregate TBD pending delta computation. At 100 residents with a 2x-per-year cadence, scaling projects to $35K-$60K annualized à la carte and $25K-$45K at partnership tier.

Scenario 3 is the operational ask. Once we can see what Invader's panel already covers, we compute the delta and either fire remaining markers through a secondary vendor or drop them from pilot scope.

Basis. Cost ranges use LabCorp/Quest à la carte rack pricing as of Apr 2026 and a <$75-per-draw partnership-tier target at volume. Vendor-confirmed rates come from direct vendor research; negotiated pricing at pilot scale will differ from the ranges shown.

Equipment and ownership. Tier 0 hardware (BP cuff, scale, tape) costs ~$65–85 per user one-time. Baseline fronts this cost; residents keep the hardware after the pilot. The wearable is a separate partnership opportunity: a subscription-model device (Whoop-style, ~$30/month, no hardware cost) fits a 4-week loanout naturally and hits sleep, HRV, activity, and HR on one device. Oura or Apple Watch work if Edge Esmeralda has existing relationships. Priority bucket is sleep. If Edge Esmeralda covers the wearable tier, that compresses ~$300 of per-pilot cost at 10 residents.

Asks, tiered

1. Lab partnership (Invader). What markers does their panel already cover? Does the partnership support week 1 and week 4 draws, or week 1 only? Are Tier 2 on-indication fires in scope, or do we route those separately?
2. Wearable partnership (optional). If Edge Esmeralda has an existing wearable partner, Baseline adapts. If not, a subscription-loanout model for the month is ~$30/user; Edge Esmeralda covering this tier compresses ~$300 of per-pilot cost at 10 residents.
3. Micro-trials sequencing (Toku Health). We understand Toku handles long-format micro-trials during the residency. We would want the week 1 and week 4 lab draws sequenced around intervention windows so Baseline scores trajectory against the intervention specifically, not just generic baseline vs follow-up. Happy to coordinate directly with Toku.
4. Everything else, Baseline fronts. BP cuffs, tapes, scales, Layer -1 intake, Baseline product, scoring, coaching infrastructure. Residents keep the hardware.