Writing
Thoughts on AI systems, ethics policy, software engineering, and building things that work.
Topics
Local LLM Benchmarks on RTX PRO 6000 Blackwell (2026): 8 Models, Real Numbers
Prompt processing, token generation, and long-context benchmarks for 8 open-weight models on an NVIDIA RTX PRO 6000 Blackwell (96 GB VRAM), measured with llama-bench at context depths up to 65K tokens.
10 min
Where Ollama Stores Your Models (And How to Use Them Directly)
Ollama stores downloaded models as plain GGUF files. Here's how to find them, use them with llama.cpp tools, and what breaks when you try.
3 min
Getting Started with NVIDIA NeMo RL + LLM
A practical guide to teaching LLMs to learn from feedback using NVIDIA NeMo RL - no PhD required.
9 min
When AI Hiring Goes Wrong
Four real cases where AI recruiting tools discriminated against candidates - and what a proper ethics audit would have caught.
7 min
Getting Started in AI Ethics Policy
You don't need a law degree to work in AI ethics. You need to understand how systems fail - and how to write the rules that prevent it.
5 min
Agentic Infrastructure: When AI Actually Does the Work
Moving from AI-assisted to AI-executed homelab management with SSH-wielding agents.
5 min
My Phone Is a Dev Environment. Here's How.
Termux, a Cloudflare tunnel, tmux, and Claude CLI. How I turned an Android phone into a real development environment connected to my home server.
4 min
Why I Built a Voice AI Platform Nobody Asked For
No pitch deck. No customers. Just a hunch that voice AI shouldn't require stitching five services together. This is how Chans.ai started - in a car, singing along to a Japanese pop song with my kids.
6 min
From Beelink to Rack: What Building a Homelab Actually Taught Me
I bought a 16-core server and a rack. Then I spent three days debugging why containers couldn't ping the internet. This is what homelabs actually teach you.
6 min
Why INGRA Failed and What I Built Instead
I wanted to give an LLM full access to my computer. Then I hit two walls I couldn't code around. One was technical. The other cost $3,000.
7 min
The Bugs That Kept Coming Back
I was hired to build frontends. Then BugCrowd reports started landing in my inbox. What I learned about security when patching stopped working.
5 min
Why I Don't Use Apple
My last Apple device was an iPhone 3GS. It stopped charging with its own cable. I never went back.
7 min
Running AI Locally Changed How I Think About Software
What happens when you stop renting GPUs and start feeling constraints. A story about Colab bills, CUDA crashes, and why friction is the best teacher.
5 min
I Didn't Climb the Tech Ladder - I Fell Into It
Opportunities expire. What two missed chances taught me about timing, readiness, and learning systems the hard way.
5 min