Local LLMs on Intel Arc: Why Running AI at Home Just Makes Sense
A cheap gaming GPU, some clever model architecture, and a coding agent that runs on local AI. Here's what I've learned hosting LLMs on an Intel Arc B580.
Essays on software, music, entrepreneurship, and building useful things from Guyana.
A cheap gaming GPU, some clever model architecture, and a coding agent that runs on local AI. Here's what I've learned hosting LLMs on an Intel Arc B580.
A journey from bash scripts on borrowed computers to building Luge Tech and self-hosting everything. The story of learning to code in Guyana's internet cafe era.
Claude and Cursor are great, but sometimes you want to build your own setup. Here's my multi-agent Pi coding system with 12 specialist agents, parallel execution, and persistent memory.
The 'vibe coder' label is becoming a slur. Here's the real divide between developers who use AI responsibly and those who don't and why the distinction matters more than the tool itself.
One wrong word can cost you $250,000 in fines. Here's how I built Lyricut.com and the Lyricut Editor to solve radio censorship problems and why people around the world is using it.
For over a decade, I've balanced software development with music production. Here's how these two worlds inform and inspire each other.
Subscribe by email or use the RSS feed for automatic updates.