Internal
Systems
OverviewPortfolioSystem Map
FoundationDoctrineIntelligence StackOperating Thesis TBDSafetyHistory LogRaw HistoryAnnual Records
JudgmentLensesDecision FrameworkJudgment Library
ExecutionPilotsSignal LogExecution SystemPrompts — SignalPrompts — FunnelPrompts — PilotPrompts — Sprint
CaptureVenturesBusiness Model TBDDistribution SystemWorkshops TBD
NetworkAllocation NetworkSpeaking & Conferences TBD
SystemsDecision DashboardFlow EngineBrand SystemBuild → AutomateKDE Engine
Systems
KDE Engine
Kallor Decision Engine · Live pipeline · Supabase
Signal → KDE → Execution → Capital
Live at kallor-decision-engine.vercel.app
Access
The Kallor Decision Engine is a separate web application running on Vercel, connected to Supabase. It processes raw signals through the three canonical AI modules and writes structured outputs to the database. Open it in a new tab to run the pipeline.
Open KDE Engine →
Three Canonical Modules Apple Notes → Supabase
01
AI Raw Signals Prompt
Parse Apple Notes and raw observations into clean signal records. Writes to raw.signals
Run →
02
AI Signal Structuring + Assessment
Structure and assess raw signals through the 6 Lenses. Writes to proc.raw_assessments
Run →
03
AI KDE Expansion Prompt
Expand assessed signals through all 6 Lenses into opportunity coordinates. Writes to proc.kde_expansions
Run →
Pipeline
Apple Notes raw.signals proc.raw_assessments proc.kde_expansions
Validate before push · Never auto-writes · Raw data is immutable
Database Supabase — Kallor Group Raw
raw
signals
Immutable ingestion layer
proc
raw_assessments
AI-structured assessments
proc
kde_expansions
6-Lens opportunity coordinates