JODIE
Organic ThemesTraditional Sector Map
Live
·Updated 2 minutes ago
About Jodie

Market structure, explained through live graph intelligence.

Jodie takes live market data and organizes it into themes, sectors, assets, alerts, and recaps so you can move from raw noise to decision-ready structure.

Organic ThemesSector Map

Model Stack

Jodie does not rely on one model pretending to know everything.
The backend runs multiple discovery experts in parallel. Correlation and eigenvectors find linear co-movement, graph communities find structural clusters, change-point logic catches breaks and regime shifts, flow models watch sponsorship and positioning, and dynamic factor models track latent loading drift. The merge layer keeps provenance intact and only upgrades a theme when the model stack agrees enough to justify it.
Ensemble
Parallel experts
Correlation/eigen, graph communities, change-point, flow, and dynamic factor models each discover themes from a different angle.
Ensemble
Quorum gate
Themes are not promoted on a single opinion. Agreement is surfaced explicitly, with confirmed, provisional, and single-expert states.
Ensemble
Provenance first
Merged themes keep source models, evidence, and agreement scores visible so every label can be audited back to the model stack.

Primary references

Correlation / eigen
Hotelling, 1933
Analysis of a Complex of Statistical Variables into Principal Components
Graph communities
Blondel et al., 2008
Fast unfolding of communities in large networks
Change-point
Killick et al., 2012
Optimal detection of changepoints with a linear computational cost
Flow driver
Pasquariello and Vega, 2007
Informed and Strategic Order Flow in the Bond Markets
Dynamic factor
Stock and Watson, 2002
Forecasting Using Principal Components From a Large Number of Predictors
01
Ingest
We pull live prices, signals, metadata, themes, and relationships from the backend pipeline.
02
Graph
Those inputs are organized into organic themes, sector relationships, propagation paths, and bridges.
03
Surface
The app turns the graph into decision screens, alerts, watchlists, and recap outputs.

How Jodie works

Lens
Organic Themes
The primary lens. This is where the market is grouped by live structure, sponsorship, breadth, and confirmation.
Lens
Traditional Sector Map
A translation layer for users who think in sectors first. It maps the same live graph into familiar language.

Daily monitoring

Track what is strengthening, what is breaking, and which names deserve attention now.

Alerts and watchlists

Save themes and assets, then watch for role changes, breadth compression, and invalidation risk.

Recaps and publishing

Turn live structure into concise market recaps, notes, and shareable commentary.

What you can do here

Navigate the live graph
Open a theme, inspect its assets, and see why it matters, what confirms it, and what breaks it.
Translate into sectors
Switch to the sector map when you need a familiar market lens for recaps, notes, or sharing with others.

FAQ

How does Jodie discover themes?
Jodie runs multiple experts in parallel. Correlation and eigenvectors, graph communities, change-point logic, flow signals, and dynamic factor models each look for themes from a different angle, then the ensemble merges the results into a single audited view.
What does the quorum gate do?
The quorum gate prevents a single model from promoting a theme on its own. Themes are tracked as confirmed, provisional, or single-expert depending on how many experts support the same latent structure.
Why does Jodie keep provenance visible?
Every merged theme keeps source models, evidence, and agreement scores visible so a user can trace a label back to the models that supported it instead of treating the output like a black box.
What is the difference between the model stack and the sector map?
The model stack is the discovery engine behind the live graph. The sector map is a translation layer that maps the same market structure into familiar sector language for quicker reading and sharing.