Context
Part of org-wide boundary planning (Linear LIM-9).
cortex-tensor provides pure-Rust tensor, transformer, and MoE building blocks (zero GPU/CUDA/Julia). It is a core math primitive that hybrid-fusion and engram-parser consumers rely on.
Goal
Document owns/does-not-own, allowed deps, and how it fits vs. SNN crates (neuromod etc.) and domain apps.
Acceptance criteria
References
- neuromod#25
- hybrid-fusion, engram-parser
Labels
documentation
Suggested branch
docs/cortex-tensor-boundary-matrix
Context
Part of org-wide boundary planning (Linear LIM-9).
cortex-tensorprovides pure-Rust tensor, transformer, and MoE building blocks (zero GPU/CUDA/Julia). It is a core math primitive that hybrid-fusion and engram-parser consumers rely on.Goal
Document owns/does-not-own, allowed deps, and how it fits vs. SNN crates (neuromod etc.) and domain apps.
Acceptance criteria
References
Labels
documentationSuggested branch
docs/cortex-tensor-boundary-matrix