Jake Goldsborough

There's a weird gap in how we describe ourselves.

Claude Code has a skill system.

A coffee roaster needed production planning software.

Last month I wrote about building an embedded comment widget for Discourse using Rust compiled to WebAssembly.

Another day, another article about AI fatigue and all the other problems it causes.

Lok hit 317 cargo installs.

Lok's spec command was useful but felt too coupled to the rest of the tool.

I've been building lok, a multi-LLM orchestration tool, and I wanted to put it through its paces on a real project.

The hardest part of building software isn't writing code.

Every developer eventually builds a setup they are proud of.

The Problem with Prompts The deeper I get into building with LLMs, the more I realize the "chat prompt" model works but could be better.

I am good with machines and bad with people.

Last time I showed lok finding 25 bugs in itself and creating GitHub issues automatically.

Part 2 ended with two promises: parallel workflow execution and dead code cleanup.

Most modern work is already mediated by machines.

Since introducing Lok, two features emerged from actual use: local LLM support via Ollama, and a declarative workflow engine.

Large language models are getting better, but they're also getting more specialized.

Intro It started with backup issues.

I wrote about using Claude Code for daily development a few months ago.

With the rise of AI and LLMs, a common theme has been "don't let the machine do the thinking".

In my previous post, I described running infrastructure with systemd services and bash scripts.

I built an embedded comment widget that lets you add Discourse-powered comments to any static site with a single script tag.

I run all my personal infrastructure on a single VPS using systemd services instead of Docker.

I wanted to track what music I listen to without sending that data to Last.fm.

2025 Yearly Recap A Year of Building, Shipping, and Showing the Work 2025 was the year things stopped being ideas in progress and started becoming systems with gravity.

I am loving CLI agentic LLM apps except for one thing... and that's losing context.

After getting laid off from my last job, I told myself I would keep my resume updated and my options open.

Snakes shed their skin.

Yesterday I published yaml-janitor using date-based version numbers like 20251115.

For the past five years, there's been an intermittent problem at work: sometimes, when editing YAML configuration files with automated tools, comments get destroyed.

I've shipped more projects in the last three months than I did in the previous year.

I live in the terminal and Neovim.

Recently, I wrote about building a Discourse plugin that visualizes invite relationships as an ASCII tree.

I built a small Discourse plugin that shows who invited whom in your community.

In Part 1, I built the core of shelltrax—a TUI music player in Rust.

We're converting one of our infrastructure services from MessageBus pub/sub to a synchronous REST API.

In Part 1, I built an Instagram-style showcase using a custom GuestSpotPost model.

I'm not exactly a social butterfly.

I've been using Claude Code for a few months now as my primary development assistant.

Part 1 covered the backend architecture, Part 2 covered topic pinning and expiration, and Part 3 covered topic boosting and custom avatar flair.

Part 1 covered the backend architecture and Part 2 covered topic pinning and the expiration system.

AWS SDK Ruby contributions and follow-up You published Contributing to AWS SDK Ruby - Part 3, which wrapped up your series about improving the Ruby AWS SDK.

Part 1 covered the backend architecture for Yaks: wallets, transactions, and the service layer.

I've wanted a random route generator for awhile now.

The Saga Continues If you've been following along, you know I've been on a journey fixing AWS SDK Ruby credential handling.

I'm building a virtual currency system for Discourse.

This week centered on publishing, polishing, and reflecting on your Discourse work, with some light data analysis and side discussions along the way.

I've been working at Discourse for a few months now, learning how flexible the platform is.

This past week marked a small but meaningful milestone for me - I submitted and merged my first pull request to the Discourse core project.

The Plot Twist Remember that AWS SDK contribution I wrote about?