Software architect building agentic AI systems — multi-agent orchestration, autonomous workflows, and the harnesses that keep long-running LLM agents on the rails. Endpoint security by day; the rest of the time I'm working out what it takes to turn Claude Code into a reliable autonomous workforce.
Featured: agentfactory
A multi-agent orchestration CLI for Claude Code. Declarative TOML workflows ("formulas"),
inter-agent mail, and crash recovery that survives context compression — turn your SKILL.md
files into autonomous agents.
af sling --agent supervisor "Fix the auth bug in login.go"- ✍️ Medium — Learn it, Live it, Share it
- 🗃️ @gstempeck — earlier work lives here


