The idea of comparing countries to choose the best one for living, working, or traveling comes to many digital nomads and remote workers. And there are already services that help with that, like the famous Numbeo. I personally love using it. However, it is crowdsourced, which means its data points …
Essential Tools for Machine Learning Engineers: My Real-World Stack
A comprehensive guide to the tools I actually use as a Machine Learning Engineer working on NLP projects. From essential development tools like PyCharm and uv, to deployment solutions like Docker and Kubernetes, to AI-powered coding assistants - this article covers the real-world toolkit that helps me ship production ML systems efficiently. Plus, honest takes on popular tools I don’t use and why.
Fitting the Player Ranking Model: A Maximum Likelihood Approach
In the previous post, we built all the necessary mathematics to model player skills as normal distributions. Here we will use all the formulas to implement the optimization algorithm in Python and fit the model to the data (both synthetic and real).
What I have learned during my 1275+ day streak on LeetCode
I started my LeetCode journey on… Well, honestly, I don’t even remember the exact date already. I had several attempts to start solving problems on LeetCode every day, but I never managed to keep it up for more than a few weeks. Until 2022, when I finally managed to …
Mathematical Model for Player Ranking
In the previous post, we discussed how to model player skills using normal distributions. Now, let’s dive into the mathematical details of how we can estimate the parameters of these distributions using Maximum Likelihood Estimation.
Problem definition: Having a set of players \(\mathcal{P}\) numbered from \(1\) to \(n …
Approximating Skills of Table Tennis Players Using Normal Distribution. Introduction
The idea struck me during a friendly table tennis tournament. We had a marathon of matches to rank ourselves by skill, and somewhere in the middle of all those games, a question popped into my mind: How can we estimate players’ skill levels \(\mathcal{S}\) using just the game outcomes …
Text classification challenge with extra-small datasets: Fine-tuning versus ChatGPT
LLMs excel on extra-small datasets, but classical approaches shine as datasets grow
Choosing the best architecture for your text classification task
A review of text classification models based on real-world applications