Search

Eval-gated CI/CD for AI products: gate merges on the confidence-interval lower bound of a bias-corrected LLM-judge score, per failure-mode axis.

Local-first multi-source video summarization with any OpenAI-compatible LLM

Attack Graph Analysis: a security toolkit for weighted directed networks (Bellman-Ford, Menger, CVSS).

Boruna: a deterministic, capability-safe programming language and framework for LLM-native applications