Senior software and data engineer based in Gdańsk, 5+ years building things in Python that run in production.
My work sits at the intersection of data engineering and AI systems — production data platforms on GCP, autonomous LLM-agent pipelines, and the backend infrastructure that ties them together. I care about correctness, testability, and shipping things that don't need babysitting.
Recent highlights:
- Built an autonomous agent platform at Morele.net handling 5,000+ weekly customer cases, automating 70%+ of complaint resolution end-to-end
- Drove dbt test coverage from ~22% to ~97% across a production catalogue serving financial reporting for a 260M+ booking marketplace
- Designed and shipped invoicing automation saving ~4 h/day of manual work; PySpark demand forecasting pipelines across ~500k SKUs per client
Data & infra: Python · SQL · dbt · Airflow · PySpark · BigQuery · GCP · Terraform · Docker
AI & automation: Claude/OpenAI APIs · autonomous agents · LangGraph · LangChain · Selenium
Engineering: CI/CD · TDD · Git · REST APIs · scikit-learn
Projects
ai-trading-signals — GPW trading signal pipeline: Selenium scrapers → LangGraph StateGraph (indicators, rules, LLM analysis) → Telegram alerts + broker execution
AirQualityDataHandler — air quality data pipeline on GCP (OpenAQ API → Cloud Function → Cloud Storage)
WedApp — wedding guest media uploader on Cloudflare Pages + D1 + Google Drive, offline-resilient upload queue
ebitda-scrapper — financial data scraper
Most of the production work is proprietary — happy to walk through it.