Graph AI — Engineering Architecture Synthesizer

Problem statement

Describe the machine or system in plain language — function, workpiece, rates, constraints.

Try:

AI settings

Anthropic API key (pre-filled with the team key; direct keyless access is used inside claude.ai)

Ontology vocabulary — composed & fully editable. Change anything, then re-synthesize.

Primary system-level function Specialization

Required function atoms

Recorded constraints (consumed by phase-B optimization)

Candidate component class per atom (swap and re-synthesize)

Feasible alternatives — pick the architecture to refine & optimize

functional (realises atom) support (port closure) environment

Phase B — one vendor product per class, exact optimization under the sizing rules below.

Objective

The full knowledge graph — fixed layout; controls restyle in place. Click a class to focus its neighborhood.

powers provides_data_to controls mounts_to mechanically_couples_to communicates_with highlight architecture