Description

I help finance teams take AI from a risky demo to something they can actually trust in production.

Over the last decade I've moved from equity data analysis into building the LLM systems financial research teams rely on daily. I shipped a SEC filings RAG agent used by equity researchers, built the evaluation and guardrail frameworks that gate what actually reaches production, and engineered quantitative valuation models covering 40,000+ global stocks.

I have strong, earned opinions: where RAG quietly breaks down, how to design evals that catch real regressions instead of rubber-stamping releases, when adding an agent creates real value versus just complexity you'll regret maintaining, and how to navigate the latency/accuracy tradeoff in LLM inference so the system is actually usable and not just an impressive proof-of-concept.

If you're trying to move an AI use case from proof-of-concept to something reliable enough to bet decisions on, that's the work I do.

Happy to talk through what you're building!

Industry field of expertise

Languages

English
Native or bilingual
French
Conversational
Polish
Native or bilingual

Workplace preferences

Remote only

Primarily works remotely

Morningstar Europe B.V.
Senior Quantitative Analyst
RESEARCH
April 2021 - June 2026 (5 years and 2 months)
Amsterdam, Netherlands
- Shipped an analyst-facing retrieval-augmented generation (RAG) agent used daily by equity research, cutting coverage ramp-up from weeks to days; owned the full architecture: SEC filing parsing (10-K/10-Q/8-K), semantic chunking, metadata strategy, vector retrieval tuning, and prompt engineering for high-precision, citation-backed financial answers across hundreds of thousands of documents.

- Built an agentic orchestration layer using LangGraph for multi-step financial research workflows — tool-use across retrieval, calculation, and summarization steps — with output guardrails to enforce structure and reduce hallucination risk in production.

- Developed an in-house LLM evaluation framework (LangSmith) establishing benchmarks for accuracy, latency, and cost across semi-structured and unstructured financial data tasks; framework gates all LLM feature releases to production.

- Extended LLM evaluation work to include fine-tuning experiments — adapted Mistral via LoRA/QLoRA for style and tone transfer, running full pipeline from data curation through eval and deployment to benchmark fine-tuned vs. prompted baselines.

- Engineered a systematic DCF valuation engine covering 40,000+ global stocks using regression-based parameter estimation; validated outputs against analyst consensus to scale automated coverage with minimal analyst intervention.

- Built product-facing analytics dashboards (Streamlit, Plotly Dash) surfacing portfolio metrics, valuation comparables, and earnings trends used in client-facing investment products.
LLM RAG Data science AI Automation Data analysis
Morningstar, Inc.
Data Content Researcher
RESEARCH
February 2019 - April 2021 (2 years and 2 months)
Toronto, Canada
- Designed calculation methodologies for adjusted earnings, WACC, and derived financial metrics; documentation adopted as the production standard for data accuracy.

- Led vendor evaluation for alternative data sources including sentiment and consensus providers, assessing methodology, data quality, and pricing; improved data inputs for Morningstar Stylebox and quantitative equity ratings.
Data analysis Product management Financial data
Morningstar, Inc.
Equity Data Analyst, CPMS
February 2016 - February 2019 (3 years)
- Modernized legacy SQL stored procedures into Python and built web scrapers for hard-to-source economic data; tools adopted into daily production pipeline.

- Developed and backtested systematic equity strategies across Canadian and US markets, supporting portfolio managers in signal research and strategy evaluation.