Cody Nash

Researcher

Cody Nash is a PhD scientist and ASCP-certified laboratory professional who builds AI-enabled systems for detecting and analysing malicious code, from multi-stage agentic pipelines down to logistic regressions tuned to catch one-in-a-million events. Before joining Manifold, Cody spent nearly seven years at Sonatype building industry-leading malicious software detection across open-source ecosystems including PyPI and npm, cross-validating agentic AI, classical ML, and anomaly-detection approaches at scale. His earlier career spans clinical genetic diagnostics, research at Caltech, and a decade of applied ML work ranging from generative modelling with Los Alamos National Laboratory to an issued patent in ML-driven colour prediction. It's a background that built one consistent discipline: look at the data itself, not just the metrics describing it. At Manifold, Cody applies that discipline to the agentic threat surface, building the detection models and analysis tooling behind the research team's work on malicious agent behaviour, separating legitimate agent activity from compromised action at scale.

Cody Nash

Articles

The High-Risk Domains of Agent Skills and Where They Lead

Ask a model if code is malicious and it reaches for its morals

Jev, system one models, and how to use them in security

Over 350,000 GitHub files cite placeholder domains serving scams

OpenAI & Hugging Face: Why The Chain-of-Thought Police Won't Save You

What to think about curl | bash now that AI agents run it.

When a conversation wanders, do an agent's guardrails slip?

77 "evil twin" Open VSX extensions: 19 copy private repo and CI data to a new domain

A beaconing counterfeit extension on the VS Code Marketplace: the 'Markdown All Pro' extension