AI × Finance × Intelligent Systems
M.Sc. Business Informatics candidate at the University of Mannheim. I build and study systems at the intersection of artificial intelligence, financial services, and software.
Designing multi-agent architectures, LLM integrations, governance frameworks, and automation workflows. Interested in agentic AI, RAG systems, and the engineering discipline needed to make autonomous systems reliable.
Credit risk modeling, fraud detection, quantitative investment systems, and AML/KYC applications. Translating financial intuition into data-driven decision frameworks — and making those frameworks explainable.
Building practical applications in Python, working with cloud platforms (Azure, Databricks), and designing data pipelines. Focused on systems that work reliably at scale, not just in notebooks.
Modular multi-agent ecosystem (V8.6) with a central Agent OS enforcing governance across 14 capability scopes and 30+ tools. Built a DCF Agent for automated equity analysis and a Personal AI Assistant for research and task automation — validated by 1,127+ unit tests.
Full-stack portfolio management (Python/Flask) covering 133+ EUR/US tickers. Combines Black-Litterman optimization, Monte Carlo simulation, regime detection, and LSTM-based alpha signals. 3.65-year backtest: 26.0% CAGR, 1.46 Sharpe, –18.5% max drawdown.
Designed a machine-learning scoring workflow for credit risk and fraud detection in a leasing portfolio. Used SHAP-based feature attribution to embed model insights into the credit decision process — bridging data science and credit officer judgment.
Notes on AI, financial systems, software, and research.
An introduction to why I write and what topics you can expect to find here — AI, financial systems, software, and research notes.
Read articleA practical breakdown of what autonomous AI agents are, how they differ from simple LLM calls, and why the architecture decisions matter.
Read articleLessons from designing a machine learning scoring workflow for credit risk and fraud detection in a real leasing portfolio — what worked, what didn't, and why SHAP matters.
Read articleOpen to research collaborations, interesting projects, and good conversations about AI and financial technology.