Artificial Intelligence & Machine Learning

This section contains my writings on artificial intelligence and machine learning, drawing from years of experience leading AI initiatives across various organizations. Topics range from practical implementation advice to theoretical discussions and industry trends.

McCabe Complexity: The Python Metric You Should Care About

April 24, 2025

Tags:python, code-quality, best-practices, development, tools

A practical guide to understanding, measuring, and managing code complexity in Python

Python Logging Best Practices for Library Developers

April 20, 2025

Tags:python, logging, libraries, best-practices, development

A comprehensive guide to implementing logging in Python libraries - from basic setup to advanced patterns and common pitfalls to avoid

Introducing 'stargazers': A Tool to Understand Your GitHub Audience

April 16, 2025

Tags:github, opensource, community, cli, python, data

Announcing a new CLI tool to fetch, analyze, and summarize the stargazers and forkers of any public GitHub repository, inspired by Cockroach Labs' analysis.

HashingEncoder: Tackling Extreme Cardinality with the Hashing Trick

April 15, 2025

Tags:python, machine-learning, category-encoders, feature-engineering, hashing-trick

Exploring the HashingEncoder - how it works, when to use it, and implementation with category_encoders

BinaryEncoder: The Space-Efficient Alternative to One-Hot Encoding

April 13, 2025

Tags:python, machine-learning, category-encoders, feature-engineering, binary-encoding

Exploring the BinaryEncoder - how it works, when to use it, and implementation with category_encoders

OrdinalEncoder: When Order Matters in Categorical Data

April 10, 2025

Tags:python, machine-learning, category-encoders, feature-engineering, ordinal-encoding

Exploring the OrdinalEncoder - how it works, when to use it, and implementation with category_encoders

Makefiles: The Unsung Hero of Python Development

April 8, 2025

Tags:python, development, tools, automation

I’ve been building Python libraries for years, and there’s one tool that consistently makes my life easier: the humble Makefile. Yes, that decades-old build automation tool from the C world. It might seem old school, but it’s become an essential part of my Python development workflow, and today I want to show you why.

What’s a Makefile Anyway?

If you’re coming from a pure Python background, you might be wondering what Make is and why we’re talking about it. Make is a build automation tool that’s been around since 1976. Think of it as a simple way to define shortcuts for common commands in your project. Instead of remembering and typing out long commands, you can define them once in a Makefile and run them with a simple make command.

Modern Python Package Publishing: PyGeoHash's New CI/CD Pipeline

April 6, 2025

Tags:python, pygeohash, github-actions, cicd, pypi, development

How we're using GitHub Actions and PyPI's Trusted Publisher to automate PyGeoHash releases

PyGeoHash Gets Type Hints: A Journey into Modern Python

April 3, 2025

Tags:python, pygeohash, type-hints, development, open-source

Adding comprehensive type hints and a new types module to PyGeoHash for better developer experience and code quality

Optimal Bankroll Management with Keeks: The Kelly Criterion

April 1, 2025

Tags:python, open-source, keeks, finance, betting, kelly-criterion

In this first post of our series on bankroll management strategies in Keeks, we’ll dive into the Kelly Criterion - the mathematical foundation of optimal betting and the inspiration behind the library’s name.

Documenting Your Library's API: Best Practices

March 30, 2025

Tags:python, documentation, api reference, sphinx, autodoc, docstrings, library, development, best practices

Create a clear, comprehensive, and easy-to-navigate API reference using Sphinx and autodoc. Learn best practices for structure, content, and cross-referencing.

OneHotEncoder: The Workhorse of Categorical Encoding

March 27, 2025

Tags:python, machine-learning, category-encoders, feature-engineering, one-hot-encoding

A deep dive into OneHotEncoder - how it works, when to use it, and implementation with category_encoders

Understanding the Elo Rating System: The Grandfather of Competitive Rankings

March 25, 2025

Tags:python, open-source, elote, rating-systems, elo, algorithms

A deep dive into the Elo rating system - how it works, its history, and implementing it with Elote

Automating Documentation Builds and Deployment with GitHub Actions and GitHub Pages

March 23, 2025

Tags:python, documentation, sphinx, github actions, ci/cd, github pages, automation, library, development

Keep your documentation up-to-date automatically. Learn how to set up a GitHub Actions workflow to build your Sphinx docs and deploy them to GitHub Pages.

Crafting Useful Code Examples: From Basic Snippets to Real-World Scenarios

March 22, 2025

Tags:python, documentation, code examples, sphinx, doctest, writing, library, development

Good code examples are crucial for documentation. Learn how to write examples that are clear, runnable, and effectively demonstrate your library's features.

Keeks 0.1.0 Release: Optimal Bankroll Management Made Simple

March 18, 2025

Tags:python, open-source, keeks, finance, betting, kelly-criterion

I’m excited to announce the release of Keeks 0.1.0, my Python library for optimal bankroll allocation and betting strategies.

If you’ve ever wondered how to mathematically optimize your betting or investment strategy, Keeks might be just what you’ve been looking for.

Getting Started with Sphinx for Python Project Documentation

March 15, 2025

Tags:python, documentation, sphinx, restructuredtext, docstrings, library, development

Generate professional, cross-referenced documentation for your Python library using Sphinx, the de facto standard tool.

Elote 1.0.0 Release: Rating Systems Made Simple

March 13, 2025

Tags:python, open-source, elote, rating-systems, elo

After what feels like forever (and honestly, it kind of has been), I’m thrilled to announce that Elote 1.0.0 has been released.

For those who haven’t been following along, Elote is my little Python library for implementing and comparing rating systems like the Elo system used in chess.

PyGeoHash v3.0.0: Faster, Freer, and More Pythonic

March 11, 2025

Tags:python, open-source, geospatial, pygeohash, performance, licensing

A deep dive into the latest major release of PyGeoHash, featuring a complete rewrite in pure CPython, MIT relicensing, and dramatic performance improvements

Using Cursor for Open Source Library Maintenance

March 9, 2025

Tags:cursor, open-source, python, ai, library-maintenance, pygeohash, keeks, elote

How AI-powered coding tools like Cursor can simplify the maintenance burden of open source projects, featuring real-world examples from PyGeoHash, Keeks, and Elote

Writing Effective Docstrings: Google vs. NumPy vs. reStructuredText Styles

March 6, 2025

Tags:python, documentation, docstrings, sphinx, autodoc, google style, numpy style, restructuredtext, library, development

Learn how to write clear, informative docstrings that Sphinx can understand, comparing the popular Google, NumPy, and native reStructuredText formats.

PyGeoHash 2.1.0: Modernizing a Geospatial Python Library

March 4, 2025

Tags:python, geospatial, open-source, geohash, cursor, claude

A look at the latest updates to PyGeoHash, a lightweight Python library for working with geohashes, and how modern AI tools helped revitalize the project.

Geohash: When Clever Isn't Always Smart

March 2, 2025

Tags:geohash, gis, python, algorithms, data-engineering

Exploring the common pitfalls and limitations of the Geohash algorithm, and when you might want to rethink using it.

Where Did All the RAM Go? Memory Profiling with Memray

March 1, 2025

Tags:python, testing, profiling, performance, memory, memray, optimization, library, development

High CPU usage isn't the only performance problem. Learn how Memray helps you track down memory leaks and excessive allocation in your Python library.

Claude 3.7 and new Cursor: first impressions

February 26, 2025

Tags:data-science, ai, tools

Anthropic finally released Claude 3.7, and Cursor has a new version with some interesting features. Let’s take a look at what’s new and some first impressions.

Finding the Slowdown: Profiling Python Code with Pyinstrument

February 25, 2025

Tags:python, testing, profiling, performance, pyinstrument, optimization, library, development

Your benchmark says a function is slow, but why? Profilers like Pyinstrument help you pinpoint exactly where your Python code is spending its time.

How Fast Is It? Benchmarking Your Code with Pytest-Benchmark

February 22, 2025

Tags:python, testing, pytest, benchmark, performance, pytest-benchmark, library, development

Functionality is crucial, but performance matters too. Learn how to easily measure the speed of your Python library code using the pytest-benchmark plugin.

From Silos to Shared Libraries: A Practical Guide to Inner Source Adoption

February 18, 2025

Tags:python, inner-source, library-development, best-practices, governance, security, development-practices

A step-by-step guide for transitioning from team-specific code to shared libraries, including governance models, security considerations, and standardized development practices.

Mastering Mocking in Python with pytest-mock

February 16, 2025

Tags:python, testing, pytest, mocking

A practical guide to mocking in Python testing - from basic concepts to advanced techniques with pytest-mock and other helpful libraries

Building Your Internal Library Developer Community

February 15, 2025

Tags:python, inner-source, community-building, corporate-culture, development-practices, library-development, best-practices

Learn how to build and nurture a thriving community of library developers within your organization through effective incentives, recognition, and collaboration practices.

Will It Blend? Testing Across Environments with Tox

February 13, 2025

Tags:python, testing, tox, pytest, library, development, compatibility, ci-cd, virtualenv

Your library works on your machine with Python 3.11. Great! But what about Python 3.9? Or 3.12? Tox ensures compatibility across different Python versions and dependency sets.

Inner Source: Bringing Open Source Culture Inside Your Organization

February 11, 2025

Tags:python, inner-source, open-source, corporate-culture, development-practices, library-development, best-practices

Learn how to harness the power of open source development practices within your organization through inner source principles and practices.

Data Science Things Roundup #13

February 10, 2025

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

Are Your Tests Enough? Measuring Coverage with Coverage.py

February 9, 2025

Tags:python, testing, pytest, coverage, pytest-cov, library, development, code-quality

Writing tests is step one. Step two is knowing what parts of your library code those tests actually exercise. Enter Coverage.py.

Designing for Developer Joy: Python Library Ergonomics

February 6, 2025

Tags:python, api-design, library-development, best-practices, developer-experience, programming, ergonomics

Explore the principles and practices that make Python libraries a joy to use, from naming conventions to error messages that guide users to solutions.

Why Your Library Needs Pytest (And How to Get Started)

February 4, 2025

Tags:python, testing, pytest, library, development, code-quality

Testing is non-negotiable for Python libraries. Let's explore why, and how the popular Pytest framework makes writing tests less painful and more powerful.

The Art of API Design: Making the Right Things Easy

February 3, 2025

Tags:python, api-design, library-development, best-practices, developer-experience, programming

Learn the principles of intuitive API design in Python, focusing on making common operations simple while keeping advanced functionality possible.

Secure Coding Practices for Python Library Developers

February 2, 2025

Tags:python, security, secure coding, best practices, library, development, input validation, least privilege, error handling

Beyond specific tools, what general principles should guide secure library development in Python? An overview of essential secure coding practices.

Taming the Python Chaos: Linting & Formatting with Ruff

January 30, 2025

Tags:python, linting, formatting, ruff, code-quality, development, ci-cd, github-actions

What linting and formatting actually are, why they matter (a lot!), and how the speedy tool Ruff can save your Python project (and your sanity).

Handling Sensitive Data Securely Within Your Python Library

January 29, 2025

Tags:python, security, sensitive data, secrets management, pii, library, development, secure coding

Does your library handle API keys, passwords, or personal information? Learn best practices for securely managing sensitive data within your Python code.

Decoding Library Updates: Understanding Semantic Versioning (SemVer)

January 28, 2025

Tags:python, packaging, versioning, semver, library, development, dependencies, pip

What does v1.2.3 actually mean? A guide for Python library authors on using Semantic Versioning to communicate changes and manage dependencies effectively.

Dependency Security: Managing Vulnerabilities with pip-audit

January 27, 2025

Tags:python, security, dependencies, vulnerabilities, pip-audit, supply chain security, library, development, safety

Your library relies on other packages. Learn how to use pip-audit to scan your dependencies for known security vulnerabilities and keep your users safe.

The Center of Your Python Project: Understanding pyproject.toml

January 26, 2025

Tags:python, packaging, pyproject.toml, pep517, pep518, pep621, setuptools, hatch, ruff, pytest, library, development

From setup.py chaos to TOML clarity. Learn why pyproject.toml exists, how it standardizes Python packaging and tool configuration, and what goes inside.

Introduction to Bandit Security Rules with Ruff: Finding Common Security Issues in Python Code

January 25, 2025

Tags:python, security, static analysis, ruff, bandit, sast, vulnerabilities, secure coding, library, development

Learn how to use Ruff's Bandit integration to automatically scan your Python code for common security pitfalls through static analysis.

Don't Forget the Fine Print: Licensing Your Python Library

January 24, 2025

Tags:python, licensing, open-source, legal, library, development, dependencies, compliance, mit, apache, gpl

Choosing an open-source license is crucial. Understand common licenses (MIT, Apache, GPL), why compatibility matters, and how to comply with obligations.

Building and Engaging a Community Around Your Open Source Library

January 22, 2025

Tags:python, open source, community, engagement, contribution, maintenance, github, library, development

An active community is vital for an open-source project's success. Learn practical steps to attract users, encourage contributions, and foster a welcoming environment.

The Library Author's Dilemma: Managing Python Dependencies

January 21, 2025

Tags:python, packaging, dependencies, library, development, pip, versions, best-practices

Choosing dependencies for your Python library is tricky. How do you decide what to include, what versions to support, and avoid creating headaches for your users?

Data Science Things Roundup #12

January 20, 2025

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

Avoiding Common Pitfalls: Injection Flaws in Python Libraries

January 18, 2025

Tags:python, security, injection, sql injection, command injection, input validation, secure coding, library, development

Injection attacks (SQL injection, command injection) aren't just for web apps. Learn how improper input handling in your Python library can create vulnerabilities.

The Art of Saying No: Defining Your Python Library's Scope

January 17, 2025

Tags:python, library, design, scope, development, software-engineering

Why keeping your Python library focused is harder than it looks, and how saying 'no' can be your most powerful design tool.

SDLC in the Age of AI

January 12, 2025

Tags:ai, software-development, programming, prompt-engineering, data science

Exploring how AI is reshaping software development practices and the emerging role of prompt engineering

From Weekend Hack to Core Tool: The category_encoders Journey

December 27, 2022

Tags:open-source, python, data science, machine-learning

Reflecting on how a simple Python experiment grew into a fundamental data science library with millions of downloads

Investment Review: Seer.ai

October 18, 2022

Tags:investing, startups, angel-investing, deal-review, ai, machine-learning, analytics

A review of my angel investment in Seer.ai, exploring how they align with my investment thesis and their unique value proposition in AI-powered analytics.

Category Encoders v1.2.8 Release

June 4, 2018

Tags:python, data science, category-encoders, open-source

Release announcement for Category Encoders v1.2.8 with bugfixes and new features

TechEmergence Podcast and Atlanta AI Article

June 3, 2018

Tags:artificial-intelligence, podcast, atlanta, predictive-maintenance, data science

Discussing predictive maintenance and the Atlanta AI ecosystem on TechEmergence

Data Engineering Podcast

February 20, 2018

Tags:data-engineering, podcast, predictive-maintenance, industrial-iot, data science

Discussing data engineering, predictive maintenance, and industrial IoT on the Data Engineering Podcast

Category Encoders published in JOSS

January 26, 2018

Tags:python, data-science, category-encoders, open-source, academic, publication

Category Encoders package gets published in the Journal of Open Source Software (JOSS)

The Problem with Industrial IoT

January 16, 2018

Tags:iot, industry, technology, data science

A discussion of the challenges facing industrial IoT adoption and implementation

Revisiting Python support in Apache Flink

January 11, 2018

Tags:python, apache-flink, big-data, streaming, data science

A follow-up on the state of Python support in Apache Flink after a couple of years

Tendencies of Data Engineers and Scientists

January 9, 2018

Tags:data-engineering, data-science, team-organization, engineering-management, data science

Exploring the relationship dynamics and organizational challenges between data engineering and data science teams

I Made a Model, Now What?

January 4, 2018

Tags:data-science, machine-learning, production, pydata, data science, atlanta

Insights from a PyData Atlanta talk about successfully deploying and maintaining machine learning models in production

On taking things too seriously: holiday edition

December 9, 2017

Tags:data science, sports, python, rating systems

A deep dive into developing rating systems for college football bowl game predictions, including the development of elote, keeks, and keeks-elote packages

Elote: a python package of rating systems

December 6, 2017

Tags:python, data-science, machine-learning, rating-systems

Introducing Elote, a Python package for implementing various rating systems

Ripyr: sampled metrics on datasets using python's asyncio

November 28, 2017

Tags:data science, python, asyncio, type hinting, data processing

An introduction to ripyr, a Python library for streaming through large datasets and parsing basic metrics using asyncio and type hinting

Category Encoders v1.2.5 Release

November 22, 2017

Tags:python, category-encoders, open-source, data-science, machine-learning

Release notes for Category Encoders v1.2.5, highlighting community contributions and improvements

Data Science Things Roundup #11

September 23, 2017

Tags:data-science, roundup, visualization, bayesian, finance

A collection of interesting data science articles and projects, including SEC keynotes, Bayesian inference, and visualization tools

git-pandas Caching: Faster Analysis

July 25, 2017

Tags:python, git, pandas, data-analysis, performance

Improving performance in git-pandas with caching

Category Encoders v1.2.4 Release

July 12, 2017

Tags:python, data-science, machine-learning, category-encoders, release

New release of category_encoders with improved functionality and bug fixes

Data Science Things Roundup #10

April 19, 2017

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

Data Science Things Roundup #9

March 12, 2017

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

Data Science Things Roundup #8

January 25, 2017

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

BaseN Encoding Grid Search in Category Encoders

December 18, 2016

Tags:python, data-science, machine-learning, category-encoders

Exploring the BaseN encoder and grid search capabilities in category_encoders

Category Encoders accepted into scikit-learn-contrib

November 20, 2016

Tags:python, data-science, open-source, scikit-learn, category-encoders

Category Encoders joins the scikit-learn-contrib ecosystem

Data Science Things Roundup #7

November 10, 2016

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

Category Encoders now on conda-forge

September 17, 2016

Tags:python, data-science, open-source, conda, category-encoders

Announcing the availability of category_encoders on conda-forge

Data Science Things Roundup #6

July 20, 2016

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

Introducing unified glob-syntax in git-pandas

June 15, 2016

Tags:git-pandas, python, data science, glob

A new, more flexible way to specify file patterns in git-pandas using glob syntax

Parallelizing cumulative blame in git-pandas with joblib

June 12, 2016

Tags:git-pandas, python, data science, performance, joblib

Improving performance of git-pandas cumulative blame analysis using parallel processing with joblib

When do I work on what?

April 30, 2016

Tags:data science, git-pandas, dataviz, oss, python

Analyzing work patterns between open source and closed source projects using git-pandas

Estimating the time spent on a project with git-pandas

April 16, 2016

Tags:git-pandas, git, github, oss, python, time tracking

Using git-pandas to estimate development time from git commit history

Data Science Things Roundup #5

March 15, 2016

Tags:data-science, machine-learning, roundup, resources

A collection of interesting data science articles and resources

Automating documentation workflow with sphinx and github pages

February 29, 2016

Tags:python, documentation, github, oss, sphinx

A guide to automating Sphinx documentation deployment to GitHub Pages

Pypi-publisher: a simple cli for publishing python libraries

February 24, 2016

Tags:python, packaging, cli, deployment, oss

Introducing pypi-publisher (ppp), a command-line tool for streamlining Python package publishing

Using survival analysis and git-pandas to estimate code quality

February 21, 2016

Tags:git-pandas, data analysis, data science, dataviz, git, github, oss, python

Using statistical survival analysis techniques with git-pandas to analyze code quality and ownership patterns

Git-pandas v1.0.0, or how to check for a stable release

February 2, 2016

Tags:git-pandas, datascience, dataviz, git, github, oss, python

Examining interface consistency and parameter naming in git-pandas for the v1.0.0 release

Github.com cumulative blame in 5 lines of python

January 31, 2016

Tags:git-pandas, data science, dataviz, git, github, pandas, python

Using git-pandas to visualize GitHub repository growth over time in just a few lines of code

Data-driven engineering team management with gitnoc and git-pandas

January 19, 2016

Tags:git-pandas, data visualization, dataviz, git, github, oss, python, software management

Using git-pandas and gitnoc to make data-driven engineering management decisions

Create organization-wide punchcards with git-pandas

January 17, 2016

Tags:git-pandas, data analysis, git, github, oss, python

Using git-pandas to create punchcard visualizations across multiple repositories

How to Write Comprehensions and Alienate People

January 8, 2016

Tags:python, programming, humor, best-practices, data science

A tongue-in-cheek guide to writing Python comprehensions that will make your colleagues question their life choices

Gitpandas v0.0.6: python 2.7, fileowners, file-wise blame and examples

January 7, 2016

Tags:git-pandas, projects, data analysis, data science, git, github, oss, python

New features and improvements in git-pandas v0.0.6 including Python 2.7 support and enhanced blame analysis

Git-Pandas v0.0.5: coverage.py, risk, and more

December 25, 2015

Tags:git-pandas, projects, data analysis, git, github, oss, pandas, python

New features in git-pandas including coverage.py support and risk analysis metrics

Common Data Pitfalls for Recurring Machine Learning Systems

December 20, 2015

Tags:data science, machine learning, analytics, data engineering

A reference guide to common data problems and limitations encountered when building recurring machine learning systems, supplementing the comprehensive guide to bad data with problems specific to recurring systems.