How far should an embodied system change after unexpected experience? Research OS, Mismatch Lab, and Builder Lab by Seonhye Gu.
What belongs in research-os under
docs/mismatch_lab/
See also PUBLIC_BOUNDARY.md
| Asset | Purpose |
|---|---|
| v0.1_spec.md | Product spec · cases · build plan |
| diff_explorer_v0.1.html | Static demo · synthetic data |
| api_schema_v0.1.json | Report JSON schema |
| benchmark_spec_v0.1.md | Public benchmark definition |
| product_overview_v0.1.md | Public problem/product/evidence summary |
| vision_narrative_v0.2.md | Long-term research framing · honest capability bounds |
| homepage_v0.1.md | Landing copy draft |
| Category | Where |
|---|---|
| Investor ask · use of funds · pricing · cap table | Private working repo |
| Design partner names · pilot customer list | Private working repo |
| Outreach scripts · LinkedIn drafts · career / relocation | Private vault |
| Named commercial product patterns (e.g. “Vendor X pattern”) in lab copy | Use generic UX framing only |
| Links to private GitHub repos | Never in research-os |
| Unredacted pilot metrics in hero copy until confirmatory | Tier B+ label only |
| Tier | May say |
|---|---|
| Demo | Synthetic / illustrative · Case 3 educates mechanism |
| B+ | Mock pilot mechanism · preliminary · not generalization |
| C | Confirmatory sim / hardware · after pre-reg freeze only |
Updated 2026-07-29