It took me 3 weeks to configure my new Framework 13 laptop, and I'm somehow ok with that.

What I like to do in and around Taipei

A stopgap for joining non-Talos distros to a Talos Kubernetes cluster

Hetzner's Object Storage service provides an S3-compatible API

Despite how popular it is, manele has always been divisive

"As an AI language model, your RAG pipeline sucks and I'm pretty sure there's information you're not giving me..."

Pairing Obisian with Taiwan's Open Source dictionary for vocabulary acquisition and practice

A sampling strategy for the GPT-4o API 2-4 frames per second requirement

How to run GPU-accelerated ML workloads with ease using NVIDIA's AI Workbench and a CUDA-enabled EC2 instance.

Trim extraneous LLM output automatically. No regex, no string parsing.