Kamil Tagowski · PhD

Connecting research and practice.

NLP, graph ML, and document AI: published at ACL and NeurIPS, shipped in real systems.

Book an intro call Explore projects

Wrocław TechPostdoctoral Researcher & AI Engineer

Finance AIFounder & Lead · research group at Wrocław Tech

Selected work

What I build

Products I build, and team projects I lead.

All projects

Flat Explorer

Live Ongoing

AI-assisted flat hunting

Scores apartment listings, summarizes visits, and compares options side by side.

Open app Author

AI-Tax

Ongoing

Legal & tax AI platform

Search, retrieval, and reasoning over large corpora of Polish legal and tax documents, built for the people who work with them daily.

Visit Research Coordinator & AI Engineer

TACTYC Ongoing AI that generates and edits game maps: vision-language models read the terrain tile by tile, generative models redraw it.

Finance AI Ongoing An applied finance-AI project in active development, building out the team and the system.

Fields of interest NLP5 Graph ML2 Legal AI2 Agentic AI2 Benchmarks2 Datasets2 Applications5

Peer-reviewed

Recent publications

Peer-reviewed work in NLP and graph representation learning.

All publications

ACL 2026 SRW Conference

2026

Factual State Discovery Benchmark: Evaluating Fact Elicitation in Polish Tax Law

M. Bystroński, K. Tagowski, D. Janiak, J. Farganus, Ł. Augustyniak, M. Kajdanowicz, T. Kajdanowicz

A benchmark for conversational fact elicitation in Polish tax law: 500 official tax-interpretation narratives decomposed into 32,874 validated atomic facts, evaluated with a discovery-through-dialogue protocol. Even the best model recovers under half the facts on hard cases after 50 turns.

AILaw 2026 Workshop

2026

Bridging AI and Law: A Scalable Multi-Agent Platform for Quantitative Legal Analytics Across Millions of Documents

Ł. Augustyniak, K. Tagowski, A. Szymczak, J. Binkowski, A. Sawczyn, M. Skibiński, D. Janiak, M. Bystroński, G. Piotrowski, M. Bernaczyk, K. Kamiński, T. Kajdanowicz

A production-scale, multi-agent platform bridging AI and legal practice, indexing 3M+ documents and 300M+ semantic vectors, with a Quantitative Legal Agent architecture for aggregation and interpretable analysis across jurisdictions.

PDF

NeurIPS 2022 Datasets & Benchmarks

2022

This is the way: designing and compiling LEPISZCZE, a comprehensive NLP benchmark for Polish

Ł. Augustyniak, K. Tagowski, A. Sawczyn, D. Janiak, R. Bartusiak, A. Szymczak, A. Janz, P. Szymański, M. Wątroba, M. Morzy, T. Kajdanowicz, M. Piasecki

A comprehensive, reproducible benchmark for Polish NLP spanning 14 diverse tasks, with a public leaderboard and pre-computed embeddings.

PDF

I build AI so that complex systems like law, finance, and information work for all of us.

Collaborate with me

Bring me a hard problem in language, graphs, or documents.

Consulting and hands-on engineering, from a quick advisory call to a production system your team can own.

Legal & Tax AI

Search, understanding, and reasoning over large legal, tax, and administrative corpora, from retrieval to agentic extraction.

AI-Tax · 3M+ documents · ACL & AILaw 2026

Applied NLP & LLM systems

RAG, extraction, and agents on your real data, with evaluation you can trust.

Flat Explorer · live product · LLM agents end-to-end

Book an intro call Start a conversation See services

30 minutes · tell me about your data and constraints