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
Brace: Application-Layer Security Testing for AI Endpoints
Brace scans AI applications for the gap between what a model will do when injected and what the application does to stop it.
- 02
Compression Is the Detector: How ByteFlow Spots Malware Without Parsing a Single Protocol
ByteFlow reads raw PCAP flow bytes and scores anomalies by how hard they are to compress, with no protocol parser in the loop.
- 03
Fenrir: Self-Hosted MDR With Published Detection Benchmarks
A self-hosted MDR/SOC platform where every alert maps to MITRE ATT&CK, detection rules live as Sigma YAML in Git, and MTTD/MTTR numbers are reproducible, not marketing.
- 04
Style Fingerprinting for AI-Generated Code Detection
SentinalAI extracts 65 lexical and AST features from source code to test whether stylometry alone can separate human-written code from AI-generated code.
- 05
llmgate: Catch LLM Regressions Before They Ship
A CLI and Python decorator that traces LLM calls to local SQLite and fails your PR when output quality drops between runs.
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.