Profile
I am an MSc Data Science candidate specializing in retrieval-augmented generation, knowledge graphs, and LLM systems. My current thesis work focuses on building a production-grade GraphRAG chatbot over a corpus of ~880 scientific papers. Previously, I spent 1.5 years at T-Mobile Czech Republic as a Data Analyst, engineering SQL pipelines and building dashboards adopted by regional sales teams. I am currently seeking ML/NLP roles where end-to-end ownership and measurable evaluation matter.
Technical Skills
- Languages: Python (scikit-learn, PyTorch, LangChain, Flask, Pydantic), SQL (Oracle, MS SQL, MariaDB), R, Java
- ML & NLP: Transformers, BGE embeddings, cross-encoder reranking, RAG / GraphRAG, Vector databases (ChromaDB), Knowledge graphs (Neo4j), MultiQuery retrieval
- LLM Ops: OpenAI / DeepSeek / Ollama APIs, prompt engineering, RAGAS evaluation, custom evaluation frameworks, agentic workflows
- MLOps & Tools: Git, Docker, Streamlit, Jupyter, Power BI, Plotly / Dash, UCloud HPC
Selected Projects
- AlgaeBot (Master's Thesis): A domain-specific GraphRAG chatbot combining a Neo4j knowledge graph with BGE vector search in ChromaDB. Built the full ingestion pipeline (PDF extraction, LLM metadata) and evaluated chunking strategies using HOPE and RAGAS frameworks.
- Vision Transformer for Plant Disease: Implemented a Vision Transformer from scratch in PyTorch, trained on 87,000 leaf images across 38 classes, achieving 99.77% validation accuracy.
- Quantitative Trading System: Built an end-to-end trading advisor using a Weibull survival model and linear programming to maximize expected return in a virtual economy, ranking in the top 0.1% of active accounts.
Professional Experience
- Data Analyst Trainee | T-Mobile Czech Republic (Apr 2023 – Aug 2024): Engineered SQL data pipelines against Oracle and MS SQL sources to consolidate revenue data. Designed Power BI dashboards adopted by regional sales teams. Served as a translation layer between business stakeholders and the analytics team.
- Administrative Assistant | Trade Fairs Brno (2020 – 2022): Performed Czech-English technical translation, competitor web scraping, and Excel-based reporting.
Education
- MSc Data Science | University of Southern Denmark, Kolding (Expected June 2026)
- BSc Applied Computer Science | Prague University of Economics and Business (2020 – 2024)