{"id":33345708,"url":"https://github.com/ranmengc137/java-rag","last_synced_at":"2026-04-12T06:31:26.296Z","repository":{"id":324819936,"uuid":"1098599600","full_name":"ranmengc137/java-rag","owner":"ranmengc137","description":"Java 17 / Spring Boot Retrieval-Augmented Generation backend with PostgreSQL + pgvector and OpenAI.","archived":false,"fork":false,"pushed_at":"2025-11-18T02:11:27.000Z","size":22,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-11-18T04:09:46.382Z","etag":null,"topics":["backend","java","llm","openai","pgvector","rag","spring-boot"],"latest_commit_sha":null,"homepage":"","language":"Java","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ranmengc137.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-11-17T22:45:58.000Z","updated_at":"2025-11-18T02:11:31.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/ranmengc137/java-rag","commit_stats":null,"previous_names":["ranmengc137/java-rag"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/ranmengc137/java-rag","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ranmengc137%2Fjava-rag","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ranmengc137%2Fjava-rag/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ranmengc137%2Fjava-rag/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ranmengc137%2Fjava-rag/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ranmengc137","download_url":"https://codeload.github.com/ranmengc137/java-rag/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ranmengc137%2Fjava-rag/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":31706764,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-12T06:22:27.080Z","status":"ssl_error","status_checked_at":"2026-04-12T06:21:52.710Z","response_time":58,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["backend","java","llm","openai","pgvector","rag","spring-boot"],"created_at":"2025-11-22T05:00:23.620Z","updated_at":"2026-04-12T06:31:26.272Z","avatar_url":"https://github.com/ranmengc137.png","language":"Java","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Java RAG Backend\n\nSpring Boot implementation of a Retrieval-Augmented Generation (RAG) backend that ingests PDFs, indexes chunk embeddings in PostgreSQL + pgvector, and answers user questions using OpenAI GPT models.\n\n## Features\n\n- PDF ingestion via REST (`/upload`) with Apache PDFBox\n- Text chunking + OpenAI embeddings\n- Vector similarity search using PostgreSQL + pgvector\n- Question answering endpoint (`/query`) with GPT-based generation\n- Clean separation of layers: controller, service, repository, config\n\n# Dev Workflow\n\n1. `docker compose up -d` (start Postgres + pgvector)\n2. `mvn spring-boot:run`\n3. Upload a PDF, then query it with curl/Postman.\n\n## Stack\n- Java 17, Spring Boot 3\n- Spring Web, Spring Data JDBC, WebClient\n- PostgreSQL 16 + pgvector\n- Apache PDFBox for text extraction\n- OpenAI Embeddings + Chat Completions APIs\n- Maven, Docker Compose\n\n## Architecture\n```\n+-----------+      +-------------+      +------------------+      +---------------+\n|  Client   | ---\u003e |  Upload API | ---\u003e |  Chunk + Embed   | ---\u003e | PostgreSQL +  |\n| (REST)    |      |  /upload    |      |  Vector Store    |      |   pgvector    |\n+-----------+      +-------------+      +------------------+      +-------+-------+\n       |                                                        ^          |\n       v                                                        |          |\n+-----------+      +-------------+      +------------------+     |   Similarity\n|  Client   | ---\u003e |  Query API  | ---\u003e |  Retriever + LLM | ----+   search\n| (REST)    |      |  /query     |      |  Prompt Builder  | -\u003e OpenAI GPT\n+-----------+      +-------------+      +------------------+\n```\n\n## Prerequisites\n1. Java 17+\n2. Maven 3.9+\n3. Docker \u0026 Docker Compose\n4. OpenAI API key with access to embeddings + chat models\n\n## Environment Variables\nSet these before running the app (e.g., in `.env` or shell profile):\n```\nexport OPENAI_API_KEY=\u003cyour-key\u003e\nexport SPRING_DATASOURCE_URL=jdbc:postgresql://localhost:5432/ragdb\nexport SPRING_DATASOURCE_USERNAME=postgres\nexport SPRING_DATASOURCE_PASSWORD=postgres\nexport SECURITY_API_KEY=\u003coptional-shared-secret-for-clients\u003e\n```\n\n## Run PostgreSQL + pgvector\n```\ncd docker\ndocker compose up -d\n```\nThis launches PostgreSQL 16 with pgvector enabled and exposed on `localhost:5432`.\n\n## Build \u0026 Run Spring Boot App\n```\ncd java-rag\nmvn clean package\nmvn spring-boot:run\n```\nThe API becomes available at `http://localhost:8080`.\n\n## Upload PDFs\n```\ncurl -X POST http://localhost:8080/upload \\\n  -H \"Authorization: Bearer $OPENAI_API_KEY\" \\\n  -F \"file=@/path/to/document.pdf\"\n```\nResponse:\n```\n{\"documentId\":\"\u003cuuid\u003e\",\"chunksStored\":42}\n```\n\n## Ask Questions\n```\ncurl -X POST http://localhost:8080/query \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"query\":\"What are the key risks?\",\"topK\":5}'\n```\nResponse:\n```\n{\n  \"answer\": \"...\",\n  \"sources\": [\n    {\"chunkIndex\":0,\"similarity\":0.82}\n  ]\n}\n```\n\n### Stream Answers (SSE)\n```\ncurl -N -X POST http://localhost:8080/query/stream \\\n  -H \"Content-Type: application/json\" \\\n  -d '{\"query\":\"Give me a short summary\",\"topK\":5}'\n```\nStreams answer tokens as Server-Sent Events for faster first-token latency.\n\n## Project Structure\n```\njava-rag/\n├── docker/docker-compose.yml\n├── pom.xml\n├── README.md\n└── src/main/java/com/randy/rag\n    ├── config\n    ├── controller\n    ├── model\n    ├── repository\n    ├── service\n    └── RagApplication.java\n```\n\n## Future Improvements\n1. **Persistent Vector Cache** – move the in-memory cache to Redis/KeyDB and add invalidation per document.\n2. **Auth Hardening** – replace shared key with JWT/OAuth and per-identity rate limits.\n3. **Chunk Metadata** – store page numbers/sections for richer citations in responses.\n4. **LLM Guardrails** – add moderation and prompt-injection filtering.\n5. **Ops** – add health/readiness probes and autoscaling tuned for streaming traffic.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Franmengc137%2Fjava-rag","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Franmengc137%2Fjava-rag","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Franmengc137%2Fjava-rag/lists"}