Contrib

Papers

I share these academic papers (pre-prints). These papers focus on research, that isn't production ready yet. Comments and feedback are much appreciated.

Computational Global Macro with AI for Risk and Portfolio Management

July 2026

https://dx.doi.org/10.2139/ssrn.7231358

This paper evaluates AI inference-guided macro-factor awareness for portfolio optimization, a new option that comes from advances in LLMs. Using LLMs for quantitative finance is promising for some use cases but needs specific features for simulation-based testing. The architecture must provide Point in Time (PiT) inference to remove model look-ahead bias.

Research on:

• Macro factors for portfolio optimizations
• Numerical AI inference with Point in Time (PiT) using open-weight models

recall_guard - recall_guard

Measured inference-without-recall — per-prompt contamination scoring for LLM outputs.

recall_guard

• Quant architectures for AI-guided portfolio optimisation (tilt percentage based)

Whitepapers

Business and capital markets content. These concepts are in production, but not 1:1. There is an always changing proprietary layer, that cannot be shared.

Survival first. Benchmarking trading algorithms under a prop-trading survival mandate

I know... no one cares about the prop trading game. Unless... you look at it like a proofing ground with its own survival rules. Like poker. Just a bit more academic, if you would allow... And no, this isn’t for the peer reviewed academic quant journals. Or for the

Aha moments for Quants and ThinkersMarius C

Other info sites

Check out my other information sites:

Because Security

InfoSec, AI, Simple

Because Security

Simple Self-hosting

/dev/norandom

Simple Self-hostingMarius C