back to homepage
Data Engineering
Pipelines, feature engineering, and evaluation tooling by Kacper Michalik for ML-ready data.
Kacper Michalik’s data engineering work spans pipelines, feature engineering, and model evaluation—using Python, Pandas, NumPy, and Matplotlib to turn raw data into something a model or a person can act on. He treats data quality and evaluation discipline as part of shipping ML-powered products, not an afterthought.
Focus
- Data preparation and feature engineering for analysis and ML
- Training / validation / test discipline and leakage awareness
- Model evaluation metrics that match the product decision
- Tooling that turns messy inputs into charts, insights, and actions
Stack
- Python
- Pandas
- NumPy
- Matplotlib
- scikit-learn
- Data pipeline patterns for ML products
In production
- O-1A Hub — retrieval corpora and evaluation loops for domain-specific AI guidance.
- Technical writing on DZone — data splits in machine learning and model evaluation metrics explained.
- Open-source DSA resource in Python — structured practice for algorithms that often sit next to data work.
Frequently Asked Questions
What does data engineering mean in Kacper Michalik’s work?
It means building the path from raw inputs to reliable features, evaluations, and product insights—pipelines and analysis tooling with Python, Pandas, and NumPy so models and humans can make better decisions.
Which product demonstrates his data engineering?
It shows up in preparing and evaluating data for AI products such as O-1A Hub, and in public writing on train/validation/test splits and evaluation metrics.
How does this connect to his AI/ML work?
Sound data engineering underpins RAG corpora, model evaluation, and MLOps. His AI/ML and data engineering pages describe the same production loop from different angles.
Related expertise
Related
