avanwyk

A structured agentic engineering workflow for building production-grade software with Claude Code, covering context management, planning, implementation, testing, code review with subagents, and commit best practices.

Enter your email Powered by Buttondown.

A look at how Java 17 stacks up against Kotlin and Scala for development teams in 2021, and an overview of popular JVM technologies.

What does Java look like in 2021?

Donations support my work and the infrastructure required to provide it.

Name Email Address Message

Implementing super-convergence for deep neural network training in Tensorflow 2 with the 1Cycle learning rate policy.

Implementing the technique in Tensorflow 2 is straightforward.

An end-to-end example of how to create your own image dataset from scratch and train a ResNet50 convolutional neural network for image classification using the FastAI library.

Instructions for installing FastAI v1 within a freshly created Anaconda virtual environment.

This post will cover getting started with FastAI v1 at the hand of tabular data.

A few years ago I came across a method for reading academic papers which I’ve kept coming back to as a reliable systematic approach to efficiently read important papers of varying complexity.

This post gives an overview of LightGBM and aims to serve as a practical reference.

Open-source Projects # Some of my public projects are available on GitHub: https://github.com/avanwyk

I will get back to you as soon as possible!

Book # Machine Learning with LightGBM and Python: A practitioner’s guide to developing production-ready machine learning systems Andrich van Wyk Papers # Analysis of activation functions for particle swarm optimised feedforward neural networks AB van Wyk,…

About Andrich van Wyk: I’m a computer scientist, AI specialist, and software architect, specialising in agentic AI, machine learning, AI systems engineering, and enterprise software.

A key concern when dealing with cyclical features is how we can encode the values such that it is clear to the deep learning algorithm that the features occur in cycles.