Writing

A Surface-Level Look at LLM Vendor Privacy

A quick overview of how AI vendors handle your data, focusing on privacy and tracking practices.

LLM Guardrails with OpenShield

Exploring OpenShield for LLM safety and compliance

From Vibe Coding to Production: Why Your Lovable.dev MVP Needs Professional Help

AI-generated apps are fantastic for validation—but terrible for scaling. Here's what happens when your MVP gets traction.

Rust and React: Tips for Full-Stack Development

How to stay productive with Rust and React using Typeshare, Garde and Mantine.

React Native (and Android Development) on Guix

React Native development on Guix, without Android Studio using Docker.

Customize Guix Shell Environment

How to set environment variables like OPENSSL_DIR within a Guix shell manifest for project-specific development environments.

Vibe Coding with Rust: The Key To Success. Not.

It works until it doesn't. My experience with AI agents and coding.

Reduced AWS Bills by 75% with Docker

How I cut AWS costs by consolidating services with Docker and HAProxy.

Automate testing with Docker Compose and Claude Code

Use Claude Code to write tests for your Docker Compose based project, and run them automatically.

Claude Code vs. Claude API: A Developer's Comparison

Claude Code vs the Claude API vs the Console: which to use, what each costs, and when a flat subscription beats pay-per-token for real dev work.

Isolate Claude Code for Rust Development with Docker

A simple setup to use Claude Code in a Docker container for Rust development with VSCode.

Claude Sonnet 4 on the Frontend with Guardrails

Using Claude 4 Sonnet to re-write a React frontend feature with guardrails.

Using LLMs to Accelerate Non-Critical Work

LLMs are great for knocking out quick scripts and one-off tools, but they’re no substitute for a careful human review when it really counts.

Run (Cursor) AppImage on Guix

Use Guix to run AppImages in a container with the necessary dependencies.

Jekyll: Publish Globally at Near-Zero Cost

How to use Jekyll to publish content globally, in multiple languages, with virtually no maintenance or vendor lock-in and at near zero-cost.

Tame a Framework Laptop with Guix

How to make a Framework laptop run quieter on Guix.

Generate ETH wallet 20x faster on React Native

How to speed up wallet generation on React Native using ethers.js and react-native-quick-crypto.

Private Ethereum Network with Besu

Goes over how to use Besu to create a private Ethereum network (on your guix machine)

GUIX on slow, embedded devices

How to use guix on a slow, embedded device, without actually running guix on it.

Bias in LLM (AI-) shampoo comparison

Using Claude 3.5 Sonnet to compare different shampoo brands by analyzing their ingredients, with prices ranging from €5 to €30 per bottle. When brand names were included in the analysis, expensive "professional" brands like Kerastase ranked highest, but when ingredients were analyzed blindly without brand names, more affordable (or less popular) options ranked better, suggesting potential bias in the LLM's initial assessment. The experiment highlights how LLMs might be influenced by brand recognition and marketing, rather than purely evaluating product ingredients and effectiveness.