Journal, whitepapers and tooling notes on audit-grade systems, security, and building things that outlive the conversation.
Whitepapers and journal entries — newest first.
What a drone CTF taught me about the boring vulnerabilities that actually get exploited, and how to close most of them before an AI finds them first.
An empirical forensic study of whether a structured, directional transmission channel links Russian state media to monitored Polish-language outlets. Over a 52-day window (24 Jul – 14 Sep 2026), 20,010 embedding vectors and 6,487 temporal pairings are audited with exact cross-lingual cosine retrieval, sub-day latency analysis of 18 verified propagation chains, and residualized time-series cross-correlation under Monte Carlo permutation and Benjamini–Hochberg FDR control.
Field tools, not products.
One self-hosted endpoint for every AI provider, with automatic failover.
Open counter-disinformation tooling for the LUSTRO Commons feed.
A local-first, pre-write gate that checks AI coding agents before they write.
A local-first supply-chain security firewall and silent daily vulnerability radar.