Three hundred patterns for engineering work you shouldn't memorize

What it does

FPF is a collection of declarative pattern languages—over 300 in the Core alone—meant to give human engineers and AI agents a common vocabulary for systems engineering, architecture, evidence, and work coordination. You describe your current situation and question, and the framework retrieves relevant patterns with constraints, checks, and stop conditions to guide the next move. It is not a software methodology or agent runtime; it is a knowledge framework applied to fields like manufacturing, robotics, energy, and healthcare technology.

The interesting bit

The framework deliberately splits cognitive labor: the AI acts as a high-bandwidth reader of the pattern corpus, citing exact sources and exposing missing evidence, while the human supplies the actual situation, stakes, constraints, and decision authority. This keeps the agent from silently inventing context, and keeps the human from having to memorize 300 interlinked transdisciplinary patterns.

Key highlights

• 300+ transdisciplinary patterns in FPF Core, plus domain-specific suites for engineering, foundational thinking (math, physics, computation, notation), and narrativization.
• Declarative, situation-driven approach: match your current question to a pattern rather than following a prescribed workflow.
• Explicitly designed for mixed human–AI teams, with clear boundaries around evidence, authority, and decision ownership.
• Already used in working projects across manufacturing, robotics, energy, and software-intensive systems, though the authors label it an “eternal alpha.”
• Framework content is published under CC BY 4.0.

Caveats

• The authors describe the project as an “eternal alpha”—already in use but continuously changing, so expect normative drift.
• The corpus is intentionally too large for humans to memorize; practical use essentially requires an AI agent to navigate and retrieve patterns.
• Using the patterns still demands relevant domain capabilities, coordination, and resources—the framework guides reasoning but does not replace engineering work.

Verdict

Worth exploring if you are an engineer-manager or systems engineer who wants structured, citable reasoning for mixed human-AI teams in complex physical domains. Skip it if you are looking for an agent framework, coding methodology, or plug-and-play automation tool.