Software — research & engines

Research & engines.

The software built for training, evaluation, spatial work, and simulation. Each engine keeps execution local, keeps authority separate from intelligence, and publishes its evidence boundaries with the code.

Engines

Built for the work, bounded on the page.

Seven public research and development repositories. Each one states what it demonstrates and what it does not; none of them runs as a hosted service.

Native training runtime

Canopy-Foundry / Neural Foundry

A native C++/CUDA training runtime with explicit model, dataset, checkpoint, device, and request-admission boundaries. Not a hosted training service.

Operators bring their own data, models, and execution environment.

Spatial authoring

Cartographer

A standalone spatial-authoring foundation: geometry, revisions, validation, generic interchange, and bounded proposal workflows.

Private adapters stay outside the public build.

Industrial operations simulation

CHAINSIGHT

A simulation-first industrial operations world for planning, governance, bounded execution, verification, replay, and reconciliation.

Not an ERP, MES, WMS, TMS, safety controller, or physical-autonomy claim.

Physical-AI research

OMNI-Q Weird Stuff Machine

Objective-to-capability planning, simulated bimanual table operations, vision, routing, and evidence receipts.

Mock mode and simulation are documented; hardware and trained-model capability remain explicit non-claims.

Optical data transport

GLYPH

An offline reference foundation for visible-pixel binary transport with manifests, hashes, codec and FEC seams.

Optical hardware, GPU execution, throughput, and conformance remain unclaimed.

Spatial-temporal reconstruction

STRATA

An offline engine for reconstructing places across time while preserving evidence, uncertainty, branches, and provenance.

Evidence and assertions remain distinct from historical truth.

Spatial sensing

SIDECHANNEL

A local-first spatial sensing explorer with deterministic simulation, freshness and privacy gates, replay, and bounded evidence receipts.

No raw-payload, persistent-identity, actuation, or causation claim is made.

Evidence discipline

A compile is not evidence.

A successful build proves the source compiles. It does not prove GPU runtime behavior, hardware parity, or production fitness. Each engine publishes what has been demonstrated and what has not; a marked gap is never promoted into a headline.

Honest gap. Repository visibility is a reviewed release boundary, not hosted access, award status, or universal capability.
Inspect the source

Every engine above is public. Read the boundary statements, then the code.