Decision teams need to know whether a forecast was justified by the information available at the time—not merely whether hindsight can explain the result.
Intelligence · v1.0.0
Dummy
Forecast the future. Remember exactly what the past knew.
Dummy takes prediction-market intelligence past hindsight theater. Point-in-time evidence, calibration, paper twins, settlement learning, and exact replay expose whether a forecast deserved confidence when it was made. The public system demonstrates formidable research machinery while keeping live order execution human-gated.
Statusv1.0.0
Primary languagePython
LicenseRepository license
AuthorityHuman owned
01 System capabilities
What it does.
- Point-in-time forecast evidence
- Paper twins and replay
- Calibration and settlement learning
- Human-gated execution boundaries
02 Client value
What this system makes possible.
Dummy is more than a standalone repository. Its architecture can be adapted to a class of real operating problems.
Evidence timing, model confidence, paper behavior, settlement truth, and execution authority must remain separate and replayable.
COMMERCIAL APPLICATIONS
- Forecast research assistants
- Calibrated risk dashboards
- Evidence-backed recommendation logs
- High-consequence approval workflows
TRANSFERABLE COMPONENTS
- Point-in-time evidence store
- Calibration engine
- Replay harness
- Settlement learning loop
03 Artifact chamber
Every claim comes with receipts.
Interrogate the source, seize the release, and verify the machinery for yourself.
The machine is open
Source
Skip the pitch. Open the public repository and interrogate the implementation directly.
Open repository01Point-in-time forecast evidence
02Paper twins and replay
03Calibration and settlement learning
04Human-gated execution boundaries
Need this architecture aimed at an operating problem of your own?
Build a governed forecasting system