samay2504/Argus ? reverse-engineered prompt

Build me a streaming risk engine that can keep up with live market data and update exposures almost instantly as new ticks come in. I want it to handle equities, crypto, and macro data, with Python doing the data fetching and orchestration, and a fast Rust core doing the math behind the scenes.

It should take in live or near live price updates, compute rolling covariance and factor betas on the fly, and give me a snapshot of risk exposure without heavy delays or a bunch of memory churn. I also want it to be easy to plug in new data sources, and to support a simple demo that pulls from common market APIs and shows the whole pipeline working end to end.

If you need to look up current docs for the Python and Rust integration pieces, go ahead and do that.