Skip to main content
Use this cookbook when your product teaches, tutors, coaches students, or personalizes learning paths. The EdTech schema keeps the same SDK calls as the general engine, but adds structured student memory such as grade level, curriculum, weak topics, learning style, exam context, and forgetting-stage review signals.

​
Enable the domain

Use the tenant domain endpoint during onboarding or from an admin settings page. 
PATCH /v1/tenant/domain-schema
Authorization: ApiKey mem_...
Content-Type: application/json

{
  "domain_schema": "edtech"
}
You can check the active domain: 
GET /v1/tenant/domain-schema
Authorization: ApiKey mem_...

​
Add a tutoring session

import os

from memoryos import Memory

mem = Memory(api_key=os.environ["MEMORYOS_API_KEY"])

result = mem.add(
    external_user_id="student_123",
    messages=[
        {
            "role": "user",
            "content": "I am in Class 10 CBSE. My boards are in March. Algebra is okay, but I keep forgetting trigonometry identities.",
        },
        {
            "role": "assistant",
            "content": "We will revise trigonometry identities with CBSE Class 10 examples and quick recall practice.",
        },
    ],
    metadata={"source": "lesson", "subject": "math"},
)

print(result.status)
MemoryOS writes generic memories and updates the EdTech overlay internally.

​
Retrieve lesson context

Call get() immediately before your model call. 
memories = mem.get(
    query="teach this student trigonometry identities",
    external_user_id="student_123",
    limit=8,
    context_max_tokens=700,
)

system_prompt = "You are a patient math tutor."
if memories.has_context:
    system_prompt = f"{system_prompt}\n\n{memories.system_prompt_addition}"
For an EdTech tenant, system_prompt_addition can include student-specific tutoring context before the normal memory bullets.

​
Build a student dashboard

Use get_edtech_profile() only when your UI needs structured profile fields. Normal tutoring calls should use get(). 
profile = mem.get_edtech_profile("student_123")

if profile:
    print(profile.grade_level)
    print(profile.board_or_curriculum)
    print(profile.exam_name, profile.exam_date)
    print(profile.weak_topics[:3])
    print(profile.forgetting_stages)
TypeScript: 
const profile = await mem.getEdTechProfile("student_123");

if (profile?.hasExamContext) {
  console.log(profile.examName, profile.examDate);
}

console.log(profile?.weakTopics.slice(0, 3));

​
Suggested product moments

Call add() after:
  • the student introduces goals, grade, curriculum, exam, or deadline
  • the tutor observes a strong topic, weak topic, misconception, or concept gap
  • the student changes explanation style or language preference
  • a lesson ends with meaningful progress or review needs
Call get() before:
  • generating the next tutoring response
  • building a practice plan
  • choosing review questions
  • preparing an exam-readiness summary

​
Conflict behavior

The EdTech router sends personal student facts back to the student for clarification and routes workspace-level facts to tenant review.
ExampleExpected route
Exam date changed from March 10 to March 15User session clarification
Student changed from Class 10 to Class 11User session clarification
Institution curriculum changedTenant review
Student prefers Hinglish explanations nowPersonal preference update

​
Data boundaries

Keep your own source-of-truth academic records in your product database. MemoryOS stores memory extracted from interactions and makes it useful for AI context. Use your product database for:
  • enrollment
  • payment status
  • official grades
  • institution rosters
  • parent or guardian records
Use MemoryOS for:
  • learning preferences
  • weak and strong topics
  • tutoring context
  • exam goals mentioned in conversation
  • review urgency and personalized context

​
Related pages