Research-Stack/5-Applications
Brandon Schneider 1280ff3580 feat: WebGPU spatial hash storage — GPU-as-database prototype
LyteNyte grid structure stored directly in GPU memory:
- 16×16×16 = 4096 cells as WebGPU storage buffer
- Compute shaders: insert, clear, neighbor, filter, sort, aggregate
- Render pipeline: instanced quads, color by voltage mode
- Zero CPU-GPU copies: data stays in GPU memory
- LyteNyte-style API: insert(), filter(), sort(), group(), aggregate()
- Parquet/Arrow-compatible export

Keyboard: 1=insert, 2=clear, 3=filter, 4=neighbor, 5=sort, 6=modes
Mouse: drag=orbit, scroll=zoom
HUD: FPS, cell count, filter matches, max density, per-mode counts

Files:
  shaders.wgsl — 6 compute + 2 render shaders
  index.html — self-contained, no build step
  grid-storage.js — LyteNyte-style GridStorage class
2026-05-30 01:51:12 -05:00
..
AppFlowy-Cloud Add EC2 recovery backup: NixOS config, AppFlowy compose/env template, credential server bootstrap, recovery guide 2026-05-18 10:44:23 -05:00
audio-dsp Track remaining source and documentation inventory 2026-05-11 22:18:31 -05:00
audit initial: sovereign research stack (consolidated, weightless, and lfs-optimized) 2026-05-04 18:11:36 -05:00
caddy-edge feat(infra): WebRTC bridge + Caddy edge config + Tailscale Funnel 2026-05-28 13:11:53 -05:00
cff Track remaining source and documentation inventory 2026-05-11 22:18:31 -05:00
cluster-dashboard feat(infra): cluster dashboard + VCN shim indentation fix 2026-05-28 01:13:54 -05:00
compression-core Track remaining source and documentation inventory 2026-05-11 22:18:31 -05:00
dashboard/spatial-hash-gpu feat: WebGPU spatial hash storage — GPU-as-database prototype 2026-05-30 01:51:12 -05:00
hutter_prize initial: sovereign research stack (consolidated, weightless, and lfs-optimized) 2026-05-04 18:11:36 -05:00
linear-native-tauri Bump the linear-tauri-cargo-minor-patch group 2026-05-20 23:03:27 -05:00
nodupe Patch NoDupe qs vulnerability 2026-05-25 20:52:25 -05:00
notion-native-tauri Bump the notion-tauri-cargo-minor-patch group 2026-05-20 23:03:27 -05:00
out/verilog feat: 12 math enhancements — Q16 LUT, braid VCN encoder, FPGA Verilog, FFT, crypto 2026-05-28 14:49:26 -05:00
parquet_compressor Bump dashmap 2026-05-20 23:03:27 -05:00
pist-scripts initial: sovereign research stack (consolidated, weightless, and lfs-optimized) 2026-05-04 18:11:36 -05:00
plugins/substack-connector Track remaining source and documentation inventory 2026-05-11 22:18:31 -05:00
scripts fix: reed_solomon_vcn decode_rs return type (tuple → bytes extraction) 2026-05-28 17:35:55 -05:00
teleport-kanban Bump the cargo group across 3 directories with 1 update 2026-05-20 23:02:48 -05:00
tests initial: sovereign research stack (consolidated, weightless, and lfs-optimized) 2026-05-04 18:11:36 -05:00
text-to-cad docs(agents): project-wide AGENTS.md audit — cross-refs, baseline, contracts 2026-05-26 22:34:46 -05:00
tools-scripts archive: remove experimental tools-scripts, scripts, and shim probes 2026-05-25 18:14:31 -05:00
webrtc-bridge feat(infra): WebRTC bridge + Caddy edge config + Tailscale Funnel 2026-05-28 13:11:53 -05:00
finance_manager.py chore: preserve working tree before secure wipe 2026-05-13 17:36:02 -05:00
README.md initial: sovereign research stack (consolidated, weightless, and lfs-optimized) 2026-05-04 18:11:36 -05:00
requirements_swarm_api.txt initial: sovereign research stack (consolidated, weightless, and lfs-optimized) 2026-05-04 18:11:36 -05:00
visualizer_service.py initial: sovereign research stack (consolidated, weightless, and lfs-optimized) 2026-05-04 18:11:36 -05:00

5-Applications

Purpose: End-to-end pipelines, integration tests, automation scripts, benchmarks, audit.

Depends on: 0 through 4

Contents (Target)

Source Destination
5-Applications/scripts/ 5-Applications/scripts/
5-Applications/tests/ 5-Applications/tests/
5-Applications/out/ 5-Applications/out/
5-Applications/audit/ 5-Applications/audit/

Scripts

  • build_manifold_graphml.py
  • export_manifold_to_obsidian.py
  • hot_swap_daemon.py
  • hot_swap_manager.py

Pipeline

  1. Generate provably hard question (Builder: ADD clock)
  2. Route via OmnidirectionalInterface (Lean)
  3. Execute via DomainModelIntegration (Lean → Python shim)
  4. Store in Google Drive topological storage (Warden: SUBTRACT clock)
  5. Hardware triumvirate integration (Judge: PAUSE clock)