AI Engineer, Graph DB

WHO WE ARE: Magic School is the premier generative AI platform for teachers. We’re just over 2 years old, and more than 7 million teachers from all over the world have joined our platform. Join a top team at a fast growing company that is working towards real social impact. Make an account and try us out at our website and connect with our passionate community on our Wall of Love.

As a Staff AI Engineer specializing in RAG, Knowledge Graphs, and Memory Systems, you’ll architect the information infrastructure that powers Magic School’s AI agents. You’ll design and build the knowledge organization, retrieval, and memory systems that determine what educational content our agents can access, how they navigate complex curriculum relationships, and how they maintain coherent understanding across extended teaching workflows serving millions of educators.

This is a high-impact IC role where you’ll define how educational knowledge is structured, indexed, embedded, and retrieved for AI consumption, mentor engineers on advanced retrieval and graph systems, and ensure our agents can reason over rich educational content with precision and reliability.

Knowledge Graph & Semantic Architecture

• Knowledge Graph Design
  :
  Architect and implement graph-based knowledge systems (Neo4j, Neptune, etc) that represent educational content relationships, standards alignments, prerequisite chains, curriculum coherence, learning progressions, and pedagogical connections. Thus enabling agents to reason over structured educational knowledge.
• Graph Schema & Ontology Development
  :
  Design and evolve ontologies and schemas for educational content, defining entity types (standards, concepts, skills, assessments), relationship semantics, and property models that support both human comprehension and AI reasoning.
• Graph
  
  RAG Implementation
  :
  Build Graph
  
  RAG systems that combine knowledge graph traversal with vector similarity, enabling agents to retrieve not just similar content but contextually connected educational materials through semantic and structural relationships.

Retrieval Pipeline Architecture

• Advanced RAG Systems
  :
  Architect and implement sophisticated retrieval-augmented generation pipelines including hybrid search (dense + sparse), multi-stage retrieval, reranking strategies, and query understanding that surface the most relevant educational content for agent reasoning.
• Embedding & Vectorization Strategy
  :
  Design and operationalize embedding pipelines for educational content, selecting and fine-tuning embedding models, implementing chunking strategies appropriate for curriculum materials, and managing vector stores at scale for fast, accurate retrieval.
• Retrieval Evaluation & Optimization
  :
  Design evaluation pipelines that measure retrieval precision, recall, MRR, and NDCG across educational content types. Continuously optimize retrieval quality through experimentation with embedding models, chunking strategies, and ranking algorithms.

Document Ingestion & Processing

• Content Ingestion Pipelines
  :
  Build robust ingestion systems that process structured (standards documents, curriculum frameworks, JSON) and unstructured (PDFs, lesson plans, textbooks) educational content, extracting entities, relationships, and metadata for knowledge base population.
• Semantic Parsing & Extraction
  :
  Implement NLP pipelines for educational content that extract key concepts, prerequisite relationships, learning objectives, and pedagogical metadata, enriching raw content with structured annotations for improved retrieval and reasoning.

Memory & Context Management

• Long-Horizon Memory Systems
  :
  Invent and operationalize memory compaction mechanisms, session state management, and cross-conversation memory patterns that allow agents to maintain coherence across extended teaching workflows while respecting token budgets.
• Context Evaluation & Monitoring
  :
  Design evaluation frameworks that measure retrieval precision, token relevance, attention allocation, and reasoning coherence as context evolves across sessions. Work with the evaluations team on detecting context degradation and retrieval failures.

Cross-Functional & Educational Domain Collaboration

• C…