Building an ML-Powered Transaction Classifier with Retraining and A/B Testing
Every month I download a CSV from my bank with all our household transactions. Each one needs a category: groceries, fuel, mortgage, subscriptions, insurance. With 200+ transactions per month across […]...
Five Architectures for Time Series Forecasting with Large Language Models
Large Language Models are increasingly being applied to time series forecasting. Not as chatbots, but as prediction engines that leverage the pattern recognition capabilities of transformer architectures to forecast numerical […]...
Local Audio Transcription with MLX Whisper and AI agent API on Apple Silicon
I attend a lot of meetings. Some are in-person, some remote, but almost all of them benefit from having a transcript and summary afterwards. Commercial transcription services work well, but […]...
Ten Terminal Tools That Changed My Data Science Workflow
Most data scientists spend their days in Jupyter notebooks and IDEs. But a surprising amount of daily friction happens in the terminal: navigating to project folders, searching through code, inspecting […]...
Omdenkkring Wageningen: Samenwerken aan Problemen
Ik ben altijd geïnteresseerd geweest in wat mensen beweegt. Wat houdt je bezig, waar loop je tegenaan, en hoe kijk je ernaar? Die vragen fascineren me, in gesprekken met vrienden, […]...
Health-Sync: Voedings- en sportdata automatisch naar Obsidian met Python
Sinds begin 2025 gebruik ik Obsidian als centraal punt voor mijn dagelijkse planning: trainingsschema’s, studieblokken, voedingsdoelen en reflecties, alles staat in gestructureerde dagnotities. Het systeem werkt goed voor plannen, maar […]...
Health-Sync: How I Automated My Nutrition and Sports Tracking with Python and Obsidian
Since early 2025 I’ve been using Obsidian as my central hub for daily planning: training schedules, study blocks, nutrition targets, and reflections all live in structured daily notes. The system […]...
The Transformative Role of Large Language Models in Time Series Forecasting
Time Series Forecasting (TSF) is crucial in data science for predicting future values from historical data. Traditional methods often struggle with complex, variable data, but Large Language Models (LLMs) offer new capabilities. LLMs excel in capturing intricate patterns and handling unstructured data like text, enhancing forecast accuracy. They are...
How to Install and Use Nano on a Synology NAS via SSH
If you’re using a Synology NAS and prefer editing files through SSH but find VIM challenging, you’re not alone! Fortunately, you can easily install nano, a more user-friendly terminal-based text editor, on your […]...
Setting Up a Python 3.10 Virtual Environment with Pyenv and PyO3 on Macbook pro Silicon M1: A Step-by-Step Guide
In preparation for the PyData Amsterdam 2024 workshop on writing Python modules in Rust, I set up my development environment to ensure smooth sailing. If you’re interested in leveraging Rust’s […]...