REFERENCE · Fund AI OS archive · not the live product← Back to The Queryable Company
§ 06 ── METHODOLOGY ── INSTITUTIONAL REASONING

Where tacit cognition
becomes executable infrastructure.

Specialist funds have methodology that lives in partner heads — fifteen years of pattern-recognition, founder-coaching instinct, evidence-rigor calibration. None of it written down. We encode it. The result is queryable, replayable, owned by you (L2), and survives partner transitions.

§ 06.1 ── ONTOLOGY HIERARCHY

Three layers. One canonical hierarchy.

The architecture is canonically defined by three layers. The relationship between them is what the manifesto names. See /substrate §05 for the L0 / L1 / L2 IP boundary that maps onto these.

CONTINUITYMETHODOLOGYInstitutional invariantL0SUBSTRATEContinuity + governanceL1MODELProbabilistic executionL2LAYER STACK · IP BOUNDARY

FIG. 01 ── METHODOLOGY · SUBSTRATE · MODEL ── L0 · L1 · L2

METHODOLOGY
Institutional invariant. Your fund's reasoning — thesis logic, scoring frameworks, evaluation heuristics, evidence weighting, decision history, LP voice calibration. What persists across everything else.
SUBSTRATE
Continuity + governance layer. The layer that preserves methodology across model evolution. Audit chain, consent gateways, evidence-mode framework, schema migration replay, customer overlay.
MODEL
Probabilistic execution engine. The current frontier LLM running the reasoning. Replaced every 6 months by the next one. The substrate makes this replacement non-destructive.

Methodology is the invariant. Models are replaceable execution layers. L0 doctrine · L1 framework · L2 instance — three IP boundaries that make the architecture work.

§ 06.2 ── OVERLAY · L2 INSTANCE

Your YAML. Your scoring. Your voice.

L0 + L1 · FUND-AGNOSTIC ARCHITECTURE11 AGENTS · AUDIT CHAIN · L8 FLOORSsegment-agnostic by designL2 · OVERLAYcustomers/[fund].yamlscoring: framework: hexframe_v1lp_voice_tiers: [...]thesis_axes: [...]kappa_floor: 0.85consent: revocable_24hSRCDILIPRFNSLPRPFSAUDL8EREGEVLVERLOADS AT BOOTSTRAP · OWNED BY YOU (L2)

FIG. 02 ── L0+L1 ARCHITECTURE · OVERLAY INJECTION AT BOOTSTRAP

LEFT · ARCHITECTURE

The Generic Architecture (L0 + L1) is fund-agnostic — the eleven agents, the audit chain, the L8 floors, the consent gateways. It's segment-agnostic by design.

RIGHT · OVERLAY

The methodology overlay (L2) — customers/[fund].yaml — is what makes it yours. It loads at system bootstrap. Every agent references it. Every audit chain entry tags its version.

Phase 0 is when we extract the overlay from partner heads. Two 90-minute methodology workshops produce the first draft. The async questionnaire fills the structured fields. The PoC validates it works against synthetic data. By Day 10, the overlay is v0.1 — your methodology in machine-readable form, owned by you (L2), versioned in git.

§ 06.3 ── DIVERGE-AND-RECONCILE

Load-bearing decisions are solved twice.

For any decision where being wrong is expensive — IC recommendations, regulatory claims, LP letter framings, override moments — the substrate solves the problem twice via independent routes. Different reasoning paths, different evidence sets, different agents. Then compares.

DECISIONROUTE A · COMPOSITEhexframe · external evidenceROUTE B · COUNTER-EVIDENCEinternal cohort · pattern matchRECONCILEAGREE → PROCEEDDISAGREE → SURFACE

FIG. 03 ── ROUTE A · ROUTE B · COMPARATOR · SURFACE-ON-DISAGREEMENT

01 · SOLVE
Primary route
Hexframe-class scoring, evidence integration, output with confidence band.
02 · SOLVE AGAIN
Independent route
Counter-evidence search, cohort pattern matching, output with confidence band.
03 · RECONCILE
Comparator
If both routes agree (within threshold), the decision proceeds with both traces logged. If they disagree: SURFACE TO PARTNERS. The substrate never silently picks.

Hidden disagreement is the most expensive failure mode in AI deployment. We refuse to allow it.

§ 06.4 ── EVIDENCE MODES

Every output labeled by trust character.

Every claim the substrate produces carries an explicit evidence mode. When a partner reviews a Hexframe score, they don't just see the number — they see what kind of trust produced it. Calibrated trust is bounded uncertainty.

0.000.250.500.751.00[verified]0.62pubmed · clinicaltrials[inferred]0.22cohort pattern · k=11[model_derived]0.12lr synthesis[partner_override]0.04P_2 sub-dim · −0.4composite6789108.177.918.42κ ≥ 0.85 · floor

FIG. 04 ── EVIDENCE-MODE BREAKDOWN · CONFIDENCE BAND · κ ≥ 0.85 FLOOR

EXAMPLE · HEXFRAME SCORE WITH EVIDENCE-MODE BREAKDOWN
# Example: Hexframe score with evidence mode breakdown
hexframe_composite: 8.17

evidence_modes:
  [verified]:
    weight: 0.62
    sources: [pubmed:34521098, clinicaltrials:NCT04812345]
  [inferred]:
    weight: 0.22
    method: cross_portfolio_pattern_match (k=11)
  [model_derived]:
    weight: 0.12
    method: lr_synthesis_of_founder_narrative
  [partner_override]:
    weight: 0.04
    note: "P_2 sub-dimension adjusted -0.4 — founder signal weak"

flagged:
  - counter_evidence: pubmed:36101234 (resolution_required)

confidence_band: [7.91, 8.42]
kappa_last_verified: 0.87  # ≥ 0.85 floor maintained

Three modes label every output: [known] from system memory, [verified] just-now via tool, [inferred] from reasoning. Calibrated. Bounded. Inspectable.

§ 06.5 ── SCORING

The Hexframe-class primitive.

Every specialist fund has some version of a multi-axis scoring framework. We refer to ours abstractly as a Hexframe-class primitive — three top-level axes, six sub-dimensions, configurable per segment. The Impact overlay realizes this as one set of axes. A Climate overlay realizes it differently. A Healthtech overlay differently still. The pattern survives; the realization adapts.

SUB_1ASUB_1BSUB_2ASUB_2BSUB_3ASUB_3BAXIS_1w · 0.40AXIS_2w · 0.30AXIS_3w · 0.30thesis_fit · 7.0diligence_pursue · 7.5ic_recommend · 8.0HEXFRAMEv1

FIG. 05 ── 3 AXES · 6 SUB-DIMENSIONS · 3 THRESHOLDS · v1

OVERLAY · SCORING SCHEMA · GIT-VERSIONED
scoring:
  framework: hexframe_v1
  axes:
    - { name: AXIS_1, sub: [SUB_1A, SUB_1B] }
    - { name: AXIS_2, sub: [SUB_2A, SUB_2B] }
    - { name: AXIS_3, sub: [SUB_3A, SUB_3B] }
  weighting:
    AXIS_1: 0.40    # configurable per overlay
    AXIS_2: 0.30
    AXIS_3: 0.30
  thresholds:
    thesis_fit:        7.0
    diligence_pursue:  7.5
    ic_recommend:      8.0
  override_conditions:
    - { trigger: founder_signal_strong, max_delta:  1.0 }
    - { trigger: contested_evidence,    max_delta: -1.5 }
  kappa_verification:
    floor:    0.85
    cadence:  weekly
    drift_response: halt_scorer_emit_blocker
§ 06.6 ── FOUNDER ENGAGEMENT

The pattern, captured.

Founder coaching is the hardest methodology to encode. Most specialist funds have a partner whose coaching style is the single most valuable thing the fund delivers to portcos. We capture that pattern as workflow, not text.

FOUNDERconsent-bounded signalSYSTEMpattern · surfaces rec.PARTNERdecides engagement01 · SIGNAL02 · PATTERN03 · SURFACEpartner-decided cadence04 · ENGAGEL8 · F2 · ABSENT

FIG. 06 ── FOUNDER · SYSTEM · PARTNER ── L8 F2 (DIRECT PATH ABSENT)

01 · DETECTION
Conversational pattern
The partner's coaching style — when to challenge, when to support, when to defer — captured as conditional workflows. Patterns triggered by founder operational signals (consent-bounded, opt-in).
02 · CADENCE
Surface vs reach-out
When the partner reaches out vs when the system surfaces a recommendation. The system never reaches out directly to founders (L8 floor F2). Recommendations surface to the partner; the partner decides whether and how to engage.
03 · TRIGGERS
Activation conditions
The conditions under which a fund's “extreme support” package gets activated. Trigger detection is signal-based; activation is partner-decided.
§ 06.7 ── TRIGGER SYSTEMS

Lead time before crisis.

Most fund failures aren't surprises. They have lead-time signals — founder communication patterns shift, calendar density spikes, sentiment proxies decline, decision velocity slows. The methodology overlay encodes which signals matter for your fund, at what thresholds, with what lead time.

≤7 days

WELLBEING SIGNAL LEAD TIME

target across all engagements

LEAD-TIME SIGNALSCRISIS EVENT≤ 7 DAYS · INTERVENTION WINDOWpartner surfaces recommendation · founder support package consideredcomm-pattern shiftT-21dcalendar density spikeT-14dsentiment proxy declineT-10ddecision velocity slowT-9dT = 0thresholds · cadence · which-signals · encoded in customers/[fund].yaml

FIG. 07 ── LEAD-TIME SIGNALS · ≤7 DAY INTERVENTION WINDOW · CRISIS

§ 06.8 ── EPISTEMIC TRUST

Calibrated trust under model uncertainty.

The sophisticated buyer's actual fear is not “the AI will hallucinate” — that's been solved enough times. The structural fear is: “How do I maintain stable confidence levels as my models get upgraded?” That's epistemic trust under model volatility. The substrate addresses it directly. LP and regulator inspection rights (D7 §12, see /substrate §13) make epistemic trust verifiable.

01 · CALIBRATED
Mode + band
Every output carries evidence mode + confidence band. Partner sees trust composition, not just trust level.
02 · BOUNDED
Floors
Confidence floors enforced at workflow level. Outputs below threshold escalate to partners automatically.
03 · DETERMINISTIC ROLLBACK
restore_to(audit_id)
restore_to(audit_id) reproduces past confidence state. Stale trust assumptions become inspectable.
04 · STABLE TOPOLOGY
Migration replay
Schema migration replay preserves historical confidence maps. Model upgrades don't invalidate past trust. D7 §12 step 4 verifies cross-tier composition integrity.
§ 06.9 ── ANTI-FRAGILITY

Every replay improves calibration.

+VALUE−VALUEYR 0YR 1YR 2YR 3YR 4YR 5SUBSTRATE · COMPOUNDSAI TOOLS · DECAY

FIG. 08 ── DECAY (TYPICAL DEPLOYMENT) · COMPOUNDING (SUBSTRATE)

LEFT · DECAY

Most AI deployments decay. Prompts age. Models churn. The marginal value of an AI tool today is lower than it was 18 months ago.

RIGHT · COMPOUNDING

The substrate compounds. Every replay test sharpens schema migration safety. Every κ ≥ 0.85 verification refines inter-rater calibration. Every Tier promotion validates the methodology overlay against observed outcomes. Every L2 → L1 pattern migration improves the framework for the whole cohort.

Every decision deepens the substrate. What you build never decays.

The model changes.
The substrate remembers.

── FUND AI OS

Methodology is the invariant. Models are replaceable execution layers. The substrate carries continuity across model volatility. That's the architecture.

§ END ── ENGAGE

If your fund has methodology worth encoding — talk to us.