About Code & Context

Hi, I'm Bryan Young. I build AI systems that have to work in production, where mistakes cost something.

I'm a Principal Software Engineer at Expel, where I build services that handle millions of security events and the APIs that put AI into analysts' workflows. Before that, I led a 12-person AI group at MITRE. I also founded Intertwine AI, where I help teams turn AI agents into workflows people can trust.

These days most of my work is secure, production AI agent systems: memory, evals, permissions, and human review. In open source, that means dspy-agent-skills, which gives coding agents tested DSPy skills, and Security Verifiers, cybersecurity RL environments with executable rewards, published on Prime Intellect's Environments Hub.

Code & Context is my field notebook. I write for engineers, founders, and technical leaders who need more than demos: what worked, what failed, what to measure, and how to keep people in control as agents take on more of the work.

Away from the keyboard, I'm a professional bassoonist.

Start here

• The night memory stopped me from doing the wrong thing: an agent, a green dry run, and the approval gate that kept it out of production.
• Introducing RL Verifiers: Teaching AI to Think Like a Security Expert: why security needs executable rewards.
• What local encrypted dotenvs protect, and when to move up: a plain threat model for secrets in agent workflows.
• Why I built dspy-agent-skills
• Building ClawFleet Academy, an ops control plane for agent fleets

Find me on GitHub and LinkedIn, or at Intertwine AI if your team wants help putting agents to work.

Subscribe if you want practical notes from the edge of AI engineering.