| Track | Owner | What Lives Here |
|---|---|---|
| Engineering layer | BIX | System architecture, data model, API specs, AI/math validation, repository structure, infrastructure design |
| Product spine | Everett | Rules and states, content logic, interaction model, component and pattern library, Calm Confidence grounded in UI |
The product layer communicates with infrastructure exclusively through documented APIs. No shared module imports, no cross-layer dependencies. This boundary is enforced from Sprint 1.
| Week | BIX Primary Output | Design Inputs Needed from Everett |
|---|---|---|
| Week 1 Mar 9-15 |
D1 scope alignment begins. D2 user flows drafted. D6 architecture skeleton begins. | Joint kickoff: division of work confirmed, prototype ownership explicit, Calm Confidence working definition aligned |
| Week 2 Mar 16-22 |
D1 draft complete. D2 draft complete. D6 skeleton complete. D3/D4 pre-loaded for technical team. | User flows and Calm Confidence framing locked by end of Week 1. BIX needs this to build architecture against real interaction patterns |
| Week 3 Mar 23-29 |
D3 complete (data model, OpenAPI specs). D6 complete. D1 and D2 iterated on client and Everett signal. | BIX provides feasibility signal on Everett prototype mid-Week 3 so architecture constraints are back to Everett before their spec locks |
| Week 4 Mar 30-Apr 5 |
D4 complete (math validation, load testing plan). D5 complete (Alpha SOW). Discovery presentation. | Component states and interaction specs locked by end of Week 3 so BIX can enter Alpha Sprint 1 without open implementation questions |
Everett owns the primary prototype. BIX uses rapid prototyping tools for feasibility validation only, testing whether a specific interaction or data flow is technically viable. BIX's schematic flow documentation (Stages 1-3, all 5 motivations) feeds Everett's prototype. There is no parallel design artifact from BIX.
The SOW's reference to an 'interactive UI prototype for Stages 1-3' is BIX's schematic flow documentation, not a high-fidelity design deliverable.
We are aligned with treating Calm Confidence as a guiding principle through Everett's prototyping and testing loop. What we want to make sure we produce together by end of Discovery is a version of that principle that is fully executable on BIX's side. Not just conceptually clear, but expressed in a form that engineers can implement from and that the system can validate against.
Specifically, by end of Week 4 we expect to have:
This is not a request to over-specify before the Alpha has been earned. It is a request that whatever Everett validates through user testing gets translated into something machines can reason about. The two outputs are complementary, and the end-of-Week 3 handshake point is where we align on the translation.
1. End of Week 1: User flows and Calm Confidence framing locked. BIX cannot finalize architecture intent for D3 and D6 without knowing what interaction patterns the system needs to support.
2. End of Week 3: Component states and interaction specs locked. Alpha Sprint 1 cannot start implementation without knowing what BIX is building toward, including the Calm Confidence criteria that will gate acceptance.
Consistent with Christopher's preference for minimal meetings: