Pi Multi-Agent Teams - Beyond the YAML Basics
Building multi-agent teams on Pi with custom extensions. Role design, communication topology, model-per-agent strategy, convergence budgets, and the TUI widget that shows where your money is going.
I help engineering teams build platform and infrastructure that keep services reliable and developers productive. I enjoy building useful systems, collaborating with thoughtful people, and untangling problems that looked much simpler on the whiteboard. Across embedded systems, on-prem, and cloud platforms, I've learned that great software comes from clear communication, practical trade-offs, and strong teams. I use this space to share ideas, lessons, and occasional reflections on engineering and leadership that might help others on a similar path.
Building multi-agent teams on Pi with custom extensions. Role design, communication topology, model-per-agent strategy, convergence budgets, and the TUI widget that shows where your money is going.
Extensions change what Pi can do. The TUI changes what Pi feels like to use. Build a custom footer with live status, a personal theme, and a prompt-picker overlay - and learn where the customization surface ends.
Skills teach the LLM what to do. Extensions teach Pi what to become. Build a safety gate, a custom slash command, and a context injector - and learn when to reach for code over markdown.
Single skills are powerful. Chained skills are transformative. Learn to compose Pi skills into multi-step workflows, bundle them into packages, and know when a skill isn't enough.
What Pi skills are, how they differ from extensions and prompts, and how teaching your agent your specific workflows makes you dramatically more efficient.