I build production AI systems and run a company-wide data + cloud platform end to end — from infrastructure and pipelines to LLM features in production.
- 🔭 Now: GenAI + a Microsoft Fabric data lake @ Agro-Sieć (sole AI/Data/Cloud engineer)
- 🛠️ Daily: Python · PyTorch · Azure · dbt · LLMs / RAG · Microsoft Fabric
- 🎓 M.Sc. in Trusted AI, Wrocław University of Science and Technology
- 🌍 Poland (Poznań area) — open to remote / hybrid / on-site
| Project | What it is |
|---|---|
| background-impact-on-classification | M.Sc. thesis — how image background affects classifiers. DeepLabV3 segmentation → ResNet50/ConvNeXt, packaged as a CLI. PyTorch |
| chess-ai | A chess engine written from scratch + a Maia neural-network opponent that plays human-like moves. Python · Pygame · TensorFlow |
| imbalanced-classification-benchmark | Benchmark of 5 class-imbalance strategies (incl. a custom UMCE ensemble) with two-way ANOVA + Tukey HSD |
| enron-spam-drift | Spam detection under concept drift: classic ML vs BERT embeddings, compared with cross-validated t-tests |
| city-trip-planning | Flask app that builds optimal city day-trips via Ant Colony Optimisation + AHP multi-criteria ranking |
| disk-forensics-toolkit | Interactive DFIR CLI: mount raw/E01 images, recover files, keyword-search PDFs, export encrypted reports |
Languages: Python · SQL/T-SQL · TypeScript · JavaScript AI / ML: LLMs (Azure OpenAI/GPT) · RAG · MCP · PyTorch · TensorFlow · scikit-learn · NLP/BERT · Computer Vision Data: Microsoft Fabric · dbt · Spark/PySpark · Power BI · BigQuery · medallion architecture Cloud & DevOps: Azure (Subscription Owner level) · Terraform/Bicep · Docker · GitHub Actions · Linux
Databricks Data Engineer Associate · Microsoft Fabric Analytics Engineer (DP-600) · Azure Data Engineer (DP-203) · Azure Fundamentals (AZ-900)



