Skip to content
Marcus Gawronsky

Projects

Selected projects

Research, software and applied AI: from original questions to systems that can be tested, reproduced and used.

Pareto frontier of candidate decisions180 candidate decisions plotted by risk, to be minimised, against return, to be maximised. The 18 non-dominated candidates, for which no alternative offers both lower risk and higher return, stay dark while the dominated majority fades. A staircase traces the efficient frontier and a ring marks one chosen trade-off near its knee.180 candidates18 non-dominatedRisk →Return ↑
Pareto frontierCandidate decisions → the non-dominated frontier → a choice

Pricing Perspective

Can company information reveal economic relationships and financial risk?

A quantitative-finance research programme connecting language-model representations, optimal transport and portfolio risk, with software for empirical tests, simulation and formal verification.

Methods: Optimal transport and Wasserstein geometry · Covariance and factor structure · Language-model representations · Reproducible research systems

Python · Lean 4 · LaTeX · Next.js

Precarious Papers

What happens when hidden financial relationships become public?

An empirical investigation of offshore entities, leaked disclosures, network structure, and market pricing, combining event studies, graph theory, and reproducible analysis.

Methods: Event studies and network econometrics

Python · Jupyter · Econometrics · Graph theory

Solenya

Search and recommendations that work from the first request.

A multimodal discovery platform connecting text and image understanding with product search and recommendations, including for visitors without interaction history.

Multimodal AI · Product discovery · Recommendation systems

insitu

Keep generated Markdown sections in sync with their source files.

A Python tool that keeps research documents connected to source files, queries and templates, with checks for stale content.

Python · Markdown · Jinja · Zensical

Talks

Explaining technical ideas without flattening them.

Presentations for research, industry, and leadership audiences on machine learning, quantitative methods, reproducibility, and organisational strategy.

HTML · Scientific communication · Data visualisation