--- title: Doing your homework with Neo4j & AI | daily.dev description: A hands-on walkthrough using Neo4j's Document Intelligence tool (intended for building GraphRAG graphs from unstructured documents) repurposed to turn a... canonical: https://daily.dev/posts/doing-your-homework-with-neo4j-ai-ebcd133bc twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Doing your homework with Neo4j & AI | daily.dev og:description: A hands-on walkthrough using Neo4j's Document Intelligence tool (intended for building GraphRAG graphs from unstructured documents) repurposed to turn a... og:url: https://daily.dev/posts/doing-your-homework-with-neo4j-ai-ebcd133bc og:image: https://api.daily.dev/og/posts/EBCD133Bc.png og:image:alt: Doing your homework with Neo4j & AI og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Doing your homework with Neo4j & AI **[neo4j](https://daily.dev/sources/neo4j)** · 15 min read · 0 upvotes · 0 comments ## Summary A hands-on walkthrough using Neo4j's Document Intelligence tool (intended for building GraphRAG graphs from unstructured documents) repurposed to turn a homework PDF into a graph of question-answer pairs for quizzing kids. The piece covers building a custom graph model in Aura Free, cleaning the imported data with Cypher, and two ways to grade answers: a no-cost Aura Agent using a Cypher Template tool, and a more advanced Java app using ai.text.structuredCompletion with an OpenAI API key (gpt-5.2) for scoring. Includes a full example test run with a history quiz on the Swedish-Danish war of 1657-1658. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I use Neo4j Document Intelligence to build a graph from a PDF document? Document Intelligence takes unstructured documents like PDFs and creates a Neo4j graph following a model you define or it proposes. In Aura, you drag a PDF into the tool, define node labels and relationships manually (or let it auto-generate a model), then click Run import to load the resulting graph into a Neo4j Aura instance, including free Aura Free instances. _Developers exploring graph-based document parsing can track Neo4j GraphRAG tooling updates on daily.dev._ ### How do I evaluate free-text answers against a correct answer using Neo4j's ai.text.structuredCompletion function? The ai.text.structuredCompletion Cypher function sends a prompt plus a JSON schema to an LLM provider like OpenAI and returns structured fields such as a numeric score and text evaluation. In practice you pass the question, the user's answer, the correct answer, and supporting context text, along with an API token and model name (for example gpt-5.2), to get back a 0-3 score and explanation. _Teams building graph-grounded AI evaluation logic can follow Cypher and GenAI plugin updates via daily.dev._ ### Can Neo4j Aura Agents write data back to the graph? No, Aura Agents currently cannot write to the graph, which limits them to read-only interactions. This means an agent-driven quiz built on Aura Agents cannot track which questions were already asked or maintain a running score, requiring a separate application layer with custom Cypher queries and state properties to handle that logic. _Developers weighing Aura Agents against custom app logic can follow Neo4j agent capability changes on daily.dev._ ## Similar posts on daily.dev - [I Built a Tiny AI Agent just for fun](https://daily.dev/posts/i-built-a-tiny-ai-agent-just-for-fun-ujqbwwcg1) · neo4j · 68 upvotes · 3 comments - [How to create AI agents with Neo4j Aura Agent](https://daily.dev/posts/how-to-create-ai-agents-with-neo4j-aura-agent-mc7jpthny) · InfoWorld · 1 upvotes · 0 comments - [Graph used right & built fast: Neo4j at HackwithBay 3.0](https://daily.dev/posts/graph-used-right-built-fast-neo4j-at-hackwithbay-3-0-zb1wg0puz) · neo4j · 2 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#neo4j](https://daily.dev/tags/neo4j) [View this post on daily.dev](https://daily.dev/posts/doing-your-homework-with-neo4j-ai-ebcd133bc) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Doing your homework with Neo4j & AI","url":"https://daily.dev/posts/doing-your-homework-with-neo4j-ai-ebcd133bc","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/doing-your-homework-with-neo4j-ai-ebcd133bc"},"datePublished":"2026-09-03T07:29:15.381Z","dateModified":"2026-09-03T09:07:44.922Z","description":"A hands-on walkthrough using Neo4j's Document Intelligence tool (intended for building GraphRAG graphs from unstructured documents) repurposed to turn a...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/2f74c16f5a254ba8db384b72c3495133?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/2f74c16f5a254ba8db384b72c3495133?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"neo4j","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"neo4j","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/a8f15865d92d4c789a415f316dcb81d2","url":"https://daily.dev/sources/neo4j"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/doing-your-homework-with-neo4j-ai-ebcd133bc","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,neo4j","timeRequired":"PT15M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"neo4j","item":"https://daily.dev/sources/neo4j"},{"@type":"ListItem","position":3,"name":"Doing your homework with Neo4j & AI"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/doing-your-homework-with-neo4j-ai-ebcd133bc#faq","mainEntity":[{"@type":"Question","name":"How can I use Neo4j Document Intelligence to build a graph from a PDF document?","acceptedAnswer":{"@type":"Answer","text":"Document Intelligence takes unstructured documents like PDFs and creates a Neo4j graph following a model you define or it proposes. In Aura, you drag a PDF into the tool, define node labels and relationships manually (or let it auto-generate a model), then click Run import to load the resulting graph into a Neo4j Aura instance, including free Aura Free instances. Developers exploring graph-based document parsing can track Neo4j GraphRAG tooling updates on daily.dev."}},{"@type":"Question","name":"How do I evaluate free-text answers against a correct answer using Neo4j's ai.text.structuredCompletion function?","acceptedAnswer":{"@type":"Answer","text":"The ai.text.structuredCompletion Cypher function sends a prompt plus a JSON schema to an LLM provider like OpenAI and returns structured fields such as a numeric score and text evaluation. In practice you pass the question, the user's answer, the correct answer, and supporting context text, along with an API token and model name (for example gpt-5.2), to get back a 0-3 score and explanation. Teams building graph-grounded AI evaluation logic can follow Cypher and GenAI plugin updates via daily.dev."}},{"@type":"Question","name":"Can Neo4j Aura Agents write data back to the graph?","acceptedAnswer":{"@type":"Answer","text":"No, Aura Agents currently cannot write to the graph, which limits them to read-only interactions. This means an agent-driven quiz built on Aura Agents cannot track which questions were already asked or maintain a running score, requiring a separate application layer with custom Cypher queries and state properties to handle that logic. Developers weighing Aura Agents against custom app logic can follow Neo4j agent capability changes on daily.dev."}}]} ```