Platform policy
Zero-spend gate, allowlist, quality floor, shared quotas.
CareerVector operatorBlueprint · careervector.blueprint-graph/v5
Graph explorer
Twenty composable top-level diagrams plus recursively nested 21×21 semantic layers. Nodes descend; boundary arrows cross to neighbours.
Cell 4.2 · product · runtime · privacy · control · feedback
Private workspace and Smart Routing
Workspace policy, feasibility, cost, and bounded learning select AI execution without leaking private content into routing evidence.
- Platform policy to Sporewright addressed field: shape root
- Workspace policy to Sporewright addressed field: shape branch
- Private D1 tensor state to Sporewright addressed field: read bounded state
- Sporewright addressed field to Ranked route + receipt: rank
- Ranked route + receipt to Local · peer · BYOK · included: dispatch within gates
- Local · peer · BYOK · included to Task pipeline: result
- Task pipeline to Outcome projector: normalized signal
- Outcome projector to Private D1 tensor state: age + project
Workspace policy
Fixed, preferred, never, BYOK, consent, curiosity, and user budget.
WorkspacePrivate D1 tensor state
Declared policy, bounded learned evidence, budget reservations; no prompts or keys.
CareerVector D1Sporewright addressed field
root → workspace → capability → stage → consumer → instance?
Open layer ↘Ranked route + receipt
Selected provider/model route, fallbacks, hard gates, and short explanation.
CareerVector routerLocal · peer · BYOK · included
Feasible executors realize options under key and privacy boundaries.
Device or serverTask pipeline
Extract, evaluate, salary, tailor, and explain return only to the workspace.
CareerVector workspaceOutcome projector
Failure, honest quality loss, latency, tokens, aging, and bounded upward messages.
CareerVector router01 · policy · as-built Platform policy Zero-spend gate, allowlist, quality floor, shared quotas.
CareerVector operator02 · policy · as-built Workspace policy Fixed, preferred, never, BYOK, consent, curiosity, and user budget.
Workspace03 · store · as-built Private D1 tensor state Declared policy, bounded learned evidence, budget reservations; no prompts or keys.
CareerVector D1- 04 · process · as-built Sporewright addressed field
root → workspace → capability → stage → consumer → instance?
Open deeper layer ↘ 05 · process · as-built Ranked route + receipt Selected provider/model route, fallbacks, hard gates, and short explanation.
CareerVector router06 · service · mixed Local · peer · BYOK · included Feasible executors realize options under key and privacy boundaries.
Device or server07 · process · mixed Task pipeline Extract, evaluate, salary, tailor, and explain return only to the workspace.
CareerVector workspace08 · process · as-built Outcome projector Failure, honest quality loss, latency, tokens, aging, and bounded upward messages.
CareerVector router
Flows
- Platform policy → Sporewright addressed field shape root
- Workspace policy → Sporewright addressed field shape branch
- Private D1 tensor state → Sporewright addressed field read bounded state
- Sporewright addressed field → Ranked route + receipt rank
- Ranked route + receipt → Local · peer · BYOK · included dispatch within gates
- Local · peer · BYOK · included → Task pipeline result
- Task pipeline → Outcome projector normalized signal
- Outcome projector → Private D1 tensor state age + project
Canonical architecture note
Open note ↗CareerVector
Product outcome
Help a person understand the market, manage opportunities, improve their materials, decide where to invest effort, and collaborate with trusted people and agents. Documents and AI are means; the private career-intelligence workspace is the product.
Authorities
- D1/Yjs owns workspace state, collaboration, CV/CL trees, notes, jobs, preferences, and application activity.
- R2 owns private workspace files and large artifacts.
- JobCache Core supplies shared public role/ad knowledge through a strict seam.
- Smart Routing selects among Included AI, consented Local AI, and user-provided AI within declared preferences, feasibility, quality, and zero-operator-cost constraints.
Invariants
- CareerVector is permanently free to users.
- Private candidate and workspace activity is not monetized.
- One workspace serves one person while supporting trusted collaboration.
- Product-local identity and private state remain below the privacy seam.
- Explicit user AI preferences are gates or policy, not suggestions for the router to erase.
- No AI success or quality is fabricated where no honest outcome signal exists.
Current state
The collaborative workspace, shared app/API/MCP foundations, document engine, job tracking, adaptive LLM routing, BYOK, and corpus seam exist in substantial form. Product experience, Local AI adoption, outcome learning, and several radar workflows continue to evolve.
Completion evidence
Human and agent journeys over the same workspace truth, offline/realtime convergence, privacy and erasure tests, honest AI-route explanation, and no-paywall product behavior.