Technical notes, experiments, theories, and build logs from my work across local AI, systems, infrastructure, and early web projects.
A physics-constrained scenario for how quantum computing could move from NISQ to fault-tolerant and distributed infrastructure without assuming its growth curve should look like AI.
A Dockerized audio and video transcription workflow with local Whisper inference, NVIDIA acceleration, CPU fallback, and privacy-focused defaults.
An in-depth look at my local AI system for durable memory, research, project work, controlled tools, voice, and evidence-backed recursive self-improvement.
A private-network systems test that routes browser chat through a VirtualBox host and isolated guest VM to a local Ollama model.
A field test of how much Instagram and TikTok content production could be automated with premade clips, including the method's peak audience results.
I built a local Docker and PowerShell bench for testing CUDA TTS voice cloning models on an RTX 5070, with notes on first run setup, model tuning, and rough inference speed.
ANNIE V2 explores memory, self-monitoring, and background cognition loops around a local model while keeping the system private and controllable.
My first local AI harness, now publicly preserved as an archival proof of concept with memory, search, vision, voice, file handling, and early code-review tools.
A February 2025 test of whether a leading LLM could develop a novel research concept and structure it as a complete paper.
The first step in building this site and the curiosity that pulled me into software, systems, and product work.
