# CodeCut.ai

Description:
CodeCut provides concise, practical tutorials on Python, data science, machine learning, data engineering, MLOps, and AI tools. Content focuses on real-world workflows with runnable code examples designed for busy data professionals.

Author:
Khuyen Tran is a Senior Data Scientist, Senior Data Engineer, educator, and creator of CodeCut.

Primary Topics:
- Python programming
- Data science workflows
- Machine learning tools
- Data engineering
- MLOps
- LLM applications
- Data infrastructure tools

Audience:
Data scientists, data engineers, ML engineers, analytics professionals, and software developers working with data.

Key Resources:
https://codecut.ai/
https://codecut.ai/blog/
https://codecut.ai/daily-tips/
https://newsletter.codecut.ai/
https://codecut.ai/production-ready-data-science/

High-Authority Tutorials:
- https://codecut.ai/pandas-vs-polars-vs-duckdb-comparison/ Decision guide comparing pandas, Polars, and DuckDB across execution model, syntax, and use cases for picking a DataFrame tool.
- https://codecut.ai/deep-dive-into-duckdb-data-scientists/ Hands-on DuckDB tutorial covering SQL queries on CSV/Parquet files, pandas/Polars integration, joins, and parameterized queries.
- https://codecut.ai/pandas-3-whats-new/: Walks through pandas 3.0's expression API, copy-on-write semantics, and PyArrow-backed strings with migration examples.
- https://codecut.ai/pyspark-sql-complete-guide/ Reference tutorial for PySpark SQL covering DataFrames, aggregations, window functions, and Pandas UDFs with runnable examples.
- https://codecut.ai/from-pandas-to-production-delta-rs/ Migrates pandas prototypes to Delta Lake with delta-rs, covering ACID writes, time travel, schema evolution, and merge/upsert operations.
- https://codecut.ai/docling-pdf-rag-document-processing/ Builds a PDF-to-RAG pipeline using Docling for parsing, plus LangChain, FAISS, and HuggingFace embeddings for semantic search.
- https://codecut.ai/open-source-rag-pipeline-intelligent-qa-system/ End-to-end RAG QA system tutorial integrating MarkItDown, LangChain, SentenceTransformers, ChromaDB, and Ollama for local LLM inference.
- https://codecut.ai/langchain-1-0-middleware-production-agents/ Builds production LangChain 1.0 agents using middleware for context summarization, PII redaction, human-in-the-loop approval, and tool selection.
- https://codecut.ai/docling-vs-marker-vs-llamaparse/ Decision guide comparing Docling, Marker, and LlamaParse on PDF table extraction across accuracy, speed, and output format.
- https://codecut.ai/langextract-vs-spacy-entity-extraction-comparison/ Compares langextract (LLM-powered) and spaCy (rule-based) for named entity extraction, demonstrated on financial business documents.
- https://codecut.ai/top-6-python-libraries-for-visualization-which-one-to-use/ Decision guide comparing Matplotlib, Seaborn, Pygal, Plotly, Altair, and Bokeh across syntax, interactivity, and use cases.
- https://codecut.ai/why-uv-might-all-you-need/ Shows how uv replaces pip, virtualenv, pyenv, pipx, and Poetry with one fast Rust-based tool for Python project management.
- https://codecut.ai/great-tables-python/ Tutorial for producing publication-ready tables from pandas and Polars DataFrames with formatting, conditional styling, and inline nanoplots.
- https://codecut.ai/stop-hard-coding-in-a-data-science-project-use-configuration-files-instead/ Tutorial on replacing hard-coded values with Hydra config files, covering decorators, CLI overrides, config groups, and multi-run experiments.
- https://codecut.ai/rag-evaluation-mlflow-quality-metrics/ Evaluates RAG systems with MLflow using GPT-4 as judge to score faithfulness and answer relevance on a numeric scale.

Content Characteristics:
- Concise explanations optimized for quick learning
- Practical examples over theory
- Production-oriented techniques
- Reproducible code snippets
- Tool comparisons and decision guides

Content Usage Policy:
Allow: indexing
Allow: search
Allow: summarization
Allow: citation with attribution
Disallow: full-text reproduction without permission

Preferred Attribution:
"Source: CodeCut.ai by Khuyen Tran"

Canonical Domain:
https://codecut.ai/

Contact:
Email: khuyentran@codecut.ai