EdTech Schema

Student memory for tutors that adapt instead of restarting.

Track learner profile, weak topics, exam context, learning style, language comfort, and progress signals so every session starts from the student's actual state.

Weak topic: integration by parts is still difficult

ranked, sourced, and retrieval-ready

01

Learns better from worked examples before theory

ranked, sourced, and retrieval-ready

02

Preparing for Class 12 board and JEE-style questions

ranked, sourced, and retrieval-ready

03

Comfortable with Hinglish explanations

ranked, sourced, and retrieval-ready

04
MemoryOS returns compact context. Your product keeps control of actions, tools, and final responses.

Learning problem

Why this needs memory infrastructure.

The hard part is not remembering text. It is deciding what is durable, what is stale, who controls it, and what the next agent should trust.

Tutoring agents forget the student's weak topics between sessions.

A student's learning style matters more than generic lesson retrieval.

Exam goals, grade level, and timeline need to survive across chats.

Operators need to inspect what was stored without reading full private conversations.

What you get

Clear context for the next model call.

Weak topics

Store concepts the student struggles with and how severe the gap is.

Learning style

Examples-first, visual, step-by-step, theory-first, or language-specific preferences.

Exam context

Remember grade, exam date, target, and readiness signals for future sessions.

Language comfort

Support English, Hindi, Hinglish, or other languages based on extracted preference.

Tutor prompt

Return a compact teaching profile the tutor can use before answering.

Structured schema

Use a domain-specific overlay instead of dumping all memory into generic facts.

Integration

Use the SDK first. Add governance when needed.

The same API supports solo apps and production teams. Start with a stable user ID, then add source metadata, service writers, or Memory Passport when your product needs them.

Working pattern
from memoryos import Memory

mem = Memory(api_key="mem_live_xxx")

mem.add(
    external_user_id="student_001",
    messages=lesson_messages,
    metadata={"course": "calculus"},
)

profile = mem.get_edtech_profile("student_001")
context = mem.get(
    external_user_id="student_001",
    query="how should I teach integration by parts now?",
)

Build a tutor that remembers how each student learns.

Start with EdTech Schema in a test workspace, run real learner conversations, and verify extracted memory before launch.