Independent research into how AI systems actually work. Published here.
I run experiments, build classifiers, and chase down the results that don't make sense. Most of what I find has been found before. Most of what I build exists somewhere in a better form. I document it anyway — the methodology, the dead ends, and the parts that didn't make it into the abstract.
- 01
EchelonAI: A Bank AI Gateway Where Compliance Is Not Optional
EchelonAI is a self-hosted AI gateway for community banks. Redaction, fair lending guardrails, and a WORM audit log sit in the request path and cannot be switched off.
- 02
Bastion: a browser where every feature is a module you can turn off
Bastion is a personal Electron browser built around a manifest-driven module system, with Mullvad routing, a fail-closed kill switch, and per-tab proxy circuits.
- 03
GRAFT: Catching Grafted Code with a Compression Model
GRAFT trains a byte-level compression model on your repo's own history and flags PR changes that don't compress like your code.
- 04
Argus NVR: One Open Port, Local Face Recognition, No Cloud Dependency
How Argus NVR proxies WebRTC camera traffic through a single port and keeps face recognition, credentials, and footage entirely on-box.
- 05
Architecture Erodes One Import at a Time
drift-scan parses TypeScript, Python, and Go imports against declared zone rules and ranks every violation by blast radius.
By day: making wooden assholes for hobby horses — things nobody asked for but someone had to build. By night: roughly the same thing, except the horses are language models and the assholes are benchmarks. Anonymous on purpose. No clout, no credentials — just the work and whether it's any good, done the way I find interesting.