Timothy Wolodzko

AI generated code and the lack of accountability

blog

LLMs make programmers “10x” more efficient, as they say. They make you produce more code in less time. By the mere scale, it is harder to quality check it. When writing code…

Collecting results from goroutines

blog

Let’s say that you are facing a problem where you have multiple jobs that need to be run, then you need to collect the results. Easy peasy, you write the code:

Forced clean, succinct code

blog

The C2 Wiki makes an interesting point

Writing parser in ocamllex and menhir

blog

OCaml is a great language for creating programming languages. In fact, it is inspired by the ML language (hence “ml” in its name), which as discussed by its author Robin…

Imitating Rust’s error propagation in Haskell

blog

Some programming languages treat errors as values that are returned, as opposed to stopping the program and showing the error message right away. The problem with stopping…

Learning Rust by implementing the lisp interpreter

blog

Rust is not the greatest language for implementing a lisp interpreter. First, lisp uses the cons linked lists for everything, while implementing them is notoriously hard in…

Hot potato state in Erlang

blog

Erlang is a peculiar language. It is a dynamically typed, functional language, with all the usual features like the preference for recursion over loops, pattern matching…

I ❤️ Quarto

blog

I recently decided to switch my page from Jekyll (GitHub Pages default) to Quatro. Both support creating pages in Markdown with configs in YAML, to build static web pages…

Lessons learned from implementing minimal Scheme four six times

blog

We were in the middle of a global pandemic. Tormented by fear, locked in our homes, everyone was coping in their own way. Some people started baking bread, writing poetry…

Pipelines: The #1 data processing design pattern

blog

In mathematics, two functions \(f\) and \(g\) can be composed \(f \circ g\), what is defined as

My favorite developer tools for Python

blog

There are many dependency management tools for python: conda, poetry, pipenv, tox, etc. Each of them has its pros and cons, fans and opponents. After spending an awful lot…

When it is easier to ask for forgiveness than permission?

blog

There are two programming styles: look before you leap (LBYL) and easier to ask for forgiveness than permission (EAFP). There are pros and cons of both styles, EAFP in some…

Makefile Programming Language Tutorial

blog

Makefile language is a functional, dynamically typed language, and of course, it is Turing complete. It’s probably the most popular, unpopular programming language, widely…

4 Ways Machine Learning Teams Use CI/CD in Production

interview

Stephen Oladele from Neptune.ai wrote an article where my work was used in one of the case studies.

Timothy Wolodzko - Staff Machine Learning Engineer @ Equinix

interview

An interview where I share my thoughts on deploying machine learning and MLOps.

Data Scientists Write Bad Code or Maybe That’s Not the Problem?

blog

Technologies and methodologies that worked for software engineering don’t necessarily do for data science. At the same time, we hear software engineers complaining about the…

Environment variables

blog

Using environment variables for storing configuration is a de facto standard. It was one of the recommendations from Heroku’s Twelve-Factor App guide and has become widely…

Can Machine Learning be Lean?

blog

Lean was a way of improving manufacturing efficiency in Toyota. Lean software development and lean startup methodologies followed it. One of the key take-aways of lean is…

MLOps at GreenSteam: Shipping Machine Learning [Case Study]

interview

I discuss the MLOps architecture I’ve build with my team at Greensteam.

Bash pocket guide

blog

Bash is like regular expressions: everyone uses it, but nobody knows it well. Every time I need to write Bash, I find myself googling a lot. The main purpose for writing…

Deploying Machine Learning Models: A Checklist

blog

In The Checklist Manifesto, Atul Gawande shows how using checklists can make everyone’s work more efficient and less error-prone. If they are useful for aircraft pilots and…

Makefiles for not-only programmers

blog

Make is commonly used in software development for managing the compilation of the source code. Use cases of make however go far beyond compiling C code, as it can be used as…

A: could someone please give a concrete example to illustrate the Dirichlet distribution prior for bag-of-words?

q&a

A: Is Bayesian Ridge Regression another name of Bayesian Linear Regression?

q&a

A: Why is the James-Stein estimator called a “shrinkage” estimator?

q&a

A: How to decide which glm family to use?

q&a

A: Best way to check implementation of density, distribution function and random generation

q&a

A: Bayes regression: how is it done in comparison to standard regression?

q&a

A: Is there any algorithm combining classification and regression?

q&a

A: Size of bootstrap samples

q&a

A: What exactly is the alpha in the Dirichlet distribution?

q&a

A: Can you explain Parzen window (kernel) density estimation in layman’s terms?

q&a

A: Do Bayesian priors become irrelevant with large sample size?

q&a

A: Fixed effect vs random effect when all possibilities are included in a mixed effects model

q&a