{"slug": "show-hn-instant-graphrag-over-any-postgres-database", "title": "Show HN: Instant GraphRAG over any Postgres database", "summary": "Evokoa launched Polygres, a managed PostgreSQL platform that integrates native graph traversal, HNSW vector search, and hybrid retrieval APIs to enable agents to query structured records, relationships, and semantic meaning from a single database. The open-source SDK and cloud service aim to simplify building AI agents that combine SQL, graph, and vector search.", "body_md": "[Evokoa](https://evokoa.com)\n\n# Postgres for the Agent Era\n\nIntroducing Polygres\n\nStructured records. Connected relationships. Semantic meaning. Polygres combines PostgreSQL hosting with native graph traversal, HNSW vector search, and hybrid retrieval APIs. Integrate using [Evokoa/polygres-sdk](https://github.com/Evokoa/polygres-sdk).\n\n## Query everything from one place.\n\nCombine SQL, graph relationships, and vector similarity search inside a single query.\n\n``` python\nfrom polygres import Polygres\n\n# Connect to the project Runtime API\nclient = Polygres(\n    api_key=\"<POLYGRES_API_KEY>\",\n    runtime_url=\"<POLYGRES_RUNTIME_URL>\",\n)\nproject = client.project()\n\n# Hybrid Search: Combine graph relationships and vector embeddings\npage = project.hybrid.joint(\n    [0.12, 0.34, 0.56],\n    {\"schema\": \"public\", \"table\": \"users\", \"id\": \"<USER_ID>\"},\n    graph_weight=0.40,\n    vector_weight=0.60,\n    limit=2\n)\n\nfor doc in page.results:\n    print(f\"[{doc.score:.4f}] {doc.properties.get('title')}\")\n{\n  \"results\": [\n    {\n      \"schema\": \"public\",\n      \"table\": \"documents\",\n      \"id\": \"doc_architecture\",\n      \"score\": 0.9634,\n      \"graph_score\": 0.92,\n      \"vector_score\": 0.99,\n      \"properties\": {\n        \"title\": \"Polygres Distributed Spec\"\n      }\n    },\n    {\n      \"schema\": \"public\",\n      \"table\": \"documents\",\n      \"id\": \"doc_pitch_deck\",\n      \"score\": 0.8415,\n      \"graph_score\": 0.69,\n      \"vector_score\": 0.94,\n      \"properties\": {\n        \"title\": \"Evokoa V2 Pitch Deck\"\n      }\n    }\n  ]\n}\n```\n\n## Everything your agent needs. Wired together.\n\nAll core components are open source. Self-host with pgGraph and pgVector, or use our managed cloud to skip the setup.\n\n### Postgres\n\nA full PostgreSQL 17 instance. Use the ORMs, clients, and tools you already know. ACID-compliant, battle-tested, and exactly what you'd expect.\n\n[View on GitHub](https://github.com/postgres/postgres)\n\n### pgGraph\n\nModel relationships between your data natively. Traverse connections, discover neighbors, and map entity graphs without wrestling with recursive SQL joins.\n\n[View on GitHub](https://www.github.com/Evokoa/pgGraph)\n\n### pgVector\n\nStore and search embeddings directly in your database. HNSW indexing for fast similarity lookups at scale, with scalar filtering to narrow results precisely.\n\n[View on GitHub](https://github.com/pgvector/pgvector)\n\n### Hybrid Search & Cloud\n\nHybrid search and Polygres scalable cloud combine Postgres, pgVector, and pgGraph together. Built with native Kubernetes and Docker scaling support, plus personalized support for pgGraph.\n\n[Create Account](https://app.polygres.com)\n\n## Your stack. Your choice.\n\nConnect with any Postgres-compatible client, use our Python SDK [Evokoa/polygres-sdk](https://github.com/Evokoa/polygres-sdk) for agent workflows, or talk directly to the REST API. Get running in minutes, not hours.\n\n``` python\nfrom polygres import Polygres\n\n# Initialize client using live retrieval key\nclient = Polygres(\n    api_key=\"<POLYGRES_API_KEY>\",\n    runtime_url=\"<POLYGRES_RUNTIME_URL>\",\n)\nproject = client.project()\n\n# Run hybrid joint retrieval matching graph proximity & vector embeddings\npage = project.hybrid.joint(\n    [0.12, 0.34, 0.56],  # Replace with your embedding vector\n    {\"schema\": \"public\", \"table\": \"users\", \"id\": \"<USER_ID>\"},\n    graph_weight=0.40,\n    vector_weight=0.60,\n    limit=5\n)\n\nfor item in page.results:\n    print(f\"[{item.score:.4f}] {item.properties.get('title')}\")\n```\n\n## Postgres for the Agent era.\n\nDeploy your managed database in seconds, or integrate with your existing schema using the open-source SDK.\n\n### Start Building with Polygres\n\nGet full access to relational table definitions, pgGraph traversals, and pgVector similarity search instantly.", "url": "https://wpnews.pro/news/show-hn-instant-graphrag-over-any-postgres-database", "canonical_source": "https://polygres.com", "published_at": "2026-07-07 21:28:59+00:00", "updated_at": "2026-07-07 22:00:20.884021+00:00", "lang": "en", "topics": ["ai-agents", "ai-infrastructure", "ai-tools"], "entities": ["Evokoa", "Polygres", "PostgreSQL", "pgGraph", "pgVector", "HNSW", "Kubernetes", "Docker"], "alternates": {"html": "https://wpnews.pro/news/show-hn-instant-graphrag-over-any-postgres-database", "markdown": "https://wpnews.pro/news/show-hn-instant-graphrag-over-any-postgres-database.md", "text": "https://wpnews.pro/news/show-hn-instant-graphrag-over-any-postgres-database.txt", "jsonld": "https://wpnews.pro/news/show-hn-instant-graphrag-over-any-postgres-database.jsonld"}}