Educator Planning Copilot
Outcome
Let a teacher turn a standards scope-and-sequence, their existing materials and the actual composition of their class section into a differentiated, standards-aligned lesson plan with aligned formative checks, without hand-searching four content repositories.
| Portfolio attribute | Value |
|---|---|
| Primary persona | Teacher |
| Primary aggregate | SectionUnitPlan |
| Cosmos DB containers | 2 |
| Grounding data | standards, resources, class profile, prior materials |
| Development estimate | USD 60-220 monthly dev estimate |
| Runtime options | Python FastAPI and C# ASP.NET Core |
Productivity hypothesis
These measures reflect the workflow, evidence, decision boundary, and industry risk for this use case.
| Impact type | Measure | Conservative internal assumption | Customer measurement |
|---|---|---|---|
| Productivity | Lesson planning time | 10-18% reduction | (baseline median planning minutes - assisted median planning minutes) / baseline median planning minutes * 100 |
| Quality or risk | Rework rate | 3-8% reduction | (baseline corrected or repeated planning-task rate - assisted rate) / baseline rate * 100 |
Important: Conservative internal pilot assumptions only; not public benchmarks, commitments, or expected customer results. Validate against a comparable baseline or matched control cohort with mandatory human review.
Rationale: Standards, resources, class profiles, and prior materials can reduce lesson research and assembly effort.
Pilot guardrail: Teachers review accuracy, accessibility, standards alignment, and instructional appropriateness.
Customer validation plan
- Baseline the current workflow with representative users and tasks.
- Run the accelerator on the same task mix with human review retained.
- Measure lesson planning time and rework rate with the documented formulas.
- Replace internal assumptions with customer-observed results before a business case.
Why Azure Cosmos DB
This AI solution needs an operational data plane, not only a vector index:
- Hybrid vector and full-text retrieval combines semantic recall with exact identifiers, codes, names, and domain language.
- Hierarchical partition keys and metadata predicates keep tenant and domain boundaries in the normal query path.
- Operational JSON, source chunks, workflow and agent state, feedback, and audit records remain together for low-latency reads and governed writes.
- Change feed reacts to source and operational updates without polling.
- Autoscale throughput, configurable consistency, and multi-region distribution support variable demand and enterprise availability.
Domain workflow

| Stage | Actor | Input → output | Approval |
|---|---|---|---|
| 1. Unit scope and pacing intake | Teacher | StandardAlignment → unitPlanSkeleton | Policy controlled |
| 2. Section composition analysis | Educator planning copilot | ClassSection → differentiationRequirements | Policy controlled |
| 3. Vetted resource retrieval | Retrieval orchestrator | differentiationRequirements → ResourceChunk | Policy controlled |
| 4. Lesson and formative check drafting | Educator planning copilot | ResourceChunk → LessonPlanDraft | Policy controlled |
| 5. Teacher revision and LMS publication | Teacher | LessonPlanDraft → publishedLesson | Human required |
Technical architecture

Cosmos DB data model
| Container | Hierarchical partition key | Documents | Search paths |
|---|---|---|---|
sectionUnitPlans | /tenantId, /sectionId | ClassSection, StandardAlignment, LessonPlanDraft, FormativeCheck | /planVectorEmbedding, /planVector |
instructionalResourceLibrary | /tenantId, /resourceDocId | ResourceChunk | /chunkVectorEmbedding, /chunkVector |
Query contract
| Query | Container | Scope | Purpose |
|---|---|---|---|
| UnitStandardsByPacing | sectionUnitPlans | single-partition | List the standards in scope for the unit in pacing order to build the lesson sequence. |
| LicenseSafeResourceRetrieval | instructionalResourceLibrary | cross-partition-bounded | Retrieve instructional resources aligned to the standard whose license permits classroom reuse and adaptation. Intentional cross-partition fan-out across the instructionalResourceLibrary reference container, bounded by the tenant prefix, the aligned standard code and the permitted license set, returning only the top-ranked resource chunks. |
| DifferentiationTierCoverage | sectionUnitPlans | single-partition | Check which lesson drafts already cover the section's below-band and English-learner tiers. |
| FormativeChecksAwaitingReview | sectionUnitPlans | single-partition | Queue generated formative items the teacher has not yet approved for administration. |
Python and C# services route each document type to its declared container and hierarchical partition key. ETag-based writes protect mutable aggregates, while append-only records and bounded cross-partition queries preserve scale.
Azure deployment

Clone the source repository, authenticate with Azure, and run:
Set-Location accelerators/education/educator-planning-copilot
az login
./Deploy.ps1 -Runtime python -Location eastus2
Change python to csharp to deploy the .NET implementation.
Explore the implementation
- Launch signed-in demo
- Implementation article
- Domain data model
- Workflow and API contract
- Python application
- C# application
- Domain container Bicep
- Platform Bicep
- Cost and configuration
- Source repository