Skip to content
View paulpel's full-sized avatar
2 followers · 2 following

Achievements

Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: Pull Sharkx2Achievement: YOLO

Achievements

Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: Pull Sharkx2Achievement: YOLO

Block or report paulpel

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
paulpel/README.md

Hi, I'm Paweł 👋

Senior AI & Data Engineer · Databricks · Azure · M.Sc. in Artificial Intelligence

I build production AI systems and the data platforms underneath them — from infrastructure and pipelines to LLM features running in production.

  • 🔭 Now: building a European data hub on Databricks @ Sii (consulting, global FMCG client)
  • 🛠️ Daily: Python · Databricks / Spark · Azure · dbt · LLMs / RAG / MCP
  • 🎓 M.Sc. in Trusted AI, Wrocław University of Science and Technology
  • 🌍 Poland (Poznań area) — open to remote / hybrid / on-site

🚀 Featured projects

Project What it is
ai-trader LLM agent platform where no LLM can emit an order — the model proposes, a deterministic RiskGate disposes. Per-agent RAG + episodic memory, three invariants enforced by code. 12-package uv workspace · 671 tests · mypy-strict · 31 ADRs
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

🧰 Tech

Languages: Python · SQL/T-SQL · TypeScript · JavaScript AI / ML: LLMs (Azure OpenAI/GPT) · RAG · MCP · PyTorch · TensorFlow · scikit-learn · NLP/BERT · Computer Vision Data: Databricks · Spark/PySpark · Microsoft Fabric · dbt · medallion architecture · Power BI · BigQuery Cloud & DevOps: Azure (Subscription Owner level) · Terraform/Bicep · Docker · GitHub Actions · Linux

📜 Certifications

Databricks Data Engineer Associate · Microsoft Fabric Analytics Engineer (DP-600) · Azure Data Engineer (DP-203) · Azure Fundamentals (AZ-900)

📫 Contact

Pinned Loading

  1. background-impact-on-classification background-impact-on-classification Public

    Study how background modifications affect image classification — ResNet50 & ConvNeXt on ImageNet animal classes (master's thesis code).

    TeX

  2. chess-ai chess-ai Public

    A chess engine built from scratch plus a Maia neural-network opponent that plays human-like moves. Python · Pygame · TensorFlow.

    Python

  3. imbalanced-classification-benchmark imbalanced-classification-benchmark Public

    Benchmarking 5 class-imbalance strategies (incl. a custom UMCE ensemble) across 3 classifiers and 12 KEEL datasets, with two-way ANOVA + Tukey HSD.

    Python

  4. enron-spam-drift enron-spam-drift Public

    Spam detection on the Enron corpus: classic ML (Naive Bayes/KNN/RF/GB) vs BERT [CLS] embeddings, under concept drift, compared with cross-validated paired t-tests.

    Python

  5. city-trip-planning city-trip-planning Public

    Flask web app that builds optimal city day-trips via Ant Colony Optimisation + AHP multi-criteria ranking, with constraint-aware scheduling and Leaflet maps.

    HTML

  6. disk-forensics-toolkit disk-forensics-toolkit Public

    Interactive CLI for digital-forensics disk-image analysis: mount raw/E01 (pytsk3/pyewf), extract metadata, recover files with foremost, keyword-search PDFs, export encrypted reports.

    Python