01
Gather point-in-time market intelligence.
We acquire and gather available information about markets, companies, sectors, analyst research, company reporting, and other point-in-time historical sources.
02
Clean and normalize diverse data.
We reconcile, timestamp, clean, and normalize the inputs into governed datasets that can support reproducible research and model development.
03
Train models to extract weak signals.
Our models learn from noisy data to identify weak market and company signals. Proprietary aggregation mechanisms combine those signals into stronger decision inputs.
04
Construct a controlled long-short book.
Portfolio rules translate ranked signals into a concentrated book of companies and ETFs, using long and limited short positions under explicit exposure controls.
05
Deploy target exposure through proprietary agents.
Our deployment agents prepare target exposures and route manager-approved order baskets to brokerage accounts through a controlled daily process.
06
Publish an explicit mark-to-market record.
Brokerage integration reconciles mark-to-market performance and supports dated publication of live and simulated records with explicit methodology and disclosure.
