PostgreSQL vs Prisma

Vector Database vs Knowledge Graph for LLMs

A practical guide comparing vector databases and knowledge graphs as storage backends for LLM applications. The article explains that vector databases (e.g., Qdrant, pgvector, Pinecone) store embeddings and excel at semantic similarity queries, while knowledge graphs (e.g., Neo4j, RDF stores) model entities and relationships and support exact, multi-hop relational queries. The author argues most real-world LLM systems need a hybrid: vectors for retrieval and graphs for precise reasoning and auditability (a pattern exemplified by Microsoft’s GraphRAG). Tool-level trade-offs, operational costs, and a four-question decision rule are provided to help teams choose the right architecture for their queries.

• Vector databases store high-dimensional embeddings and are optimized for semantic similarity (examples: Qdrant, pgvector, Pinecone, Milvus, Weaviate, Chroma).
• Knowledge graphs store entities and relationships as first-class citizens and support exact traversals and multi-hop queries using languages like Cypher or SPARQL (examples: Neo4j, RDF stores).
• The author has deployed both approaches in production: vector solutions (Qdrant and pgvector) and Neo4j for graph workloads.

Ship a RAG Chatbot with Claude, pgvector, FastAPI

This technical how-to shows how to build a retrieval-augmented generation (RAG) chatbot in a weekend using three components: PostgreSQL with the pgvector extension as the vector store, FastAPI as a thin web layer, and Anthropic's Claude for text generation. Claude does not provide an embeddings endpoint, so an external embedding provider is required (examples used are Voyage AI, OpenAI, or local sentence-transformers). The guide explains practical schema choices (embedding dimension must match pgvector column), ingestion and chunking, HNSW indexing for fast nearest-neighbor search, and the single-SQL retrieval pattern (ORDER BY embedding <=> query LIMIT k). It also emphasizes grounding via a strict system prompt and lists production hardening tasks (connection pooling, evaluation, better chunking, streaming, and auth) that should follow the initial weekend prototype.

• Postgres with the pgvector extension can serve as a vector store; create a vector column sized to the embedding model and use an HNSW index for fast nearest-neighbor search.
• Anthropic's Claude (generation) does not provide an embeddings API; the author uses an external embedding provider (Voyage AI's voyage-3, OpenAI, or local models).
• Retrieval is implemented with a single SQL query using cosine distance: ORDER BY embedding <=> query LIMIT k to get top-k chunks.

Prisma ORM Manifesto 2026: Open, Extensible, Collaborative

The five principles behind Prisma ORM 8: an open framework and ecosystem, tools that are safe to delegate, exceptional DX, modern JavaScript, and stewardship.

Prisma Is Building the Stack for the Next Million Products

Prisma is building a software factory: one environment where a builder describes what they want and an agent builds it, runs it, inspects it, fixes it, and keeps going. Prisma ORM, Prisma Postgres, and Prisma Compute, connected into one loop.

Atomicity fix for Formbricks AI demo

An open-source contributor found a bug in Formbricks' AI example-response generator where a mid-batch failure left partially committed synthetic survey data, preventing retries. The author reproduced the failure deterministically, implemented a transactional patch using Prisma transactions and a PostgreSQL FOR NO KEY UPDATE lock, and added tests to ensure generated batches either fully commit or fully roll back for the tested failure modes. The model call remains outside the transaction; state is revalidated and a narrow persistence context is used before committing. The upstream issue remained open and the proposed fix lives in the author's fork as of 2026-08-21.

• Formbricks is an open-source experience-management platform that can generate example survey responses for demo purposes.
• A bug allowed partial database commits during example-response generation: some Responses, Displays and Tag links could persist even when the overall operation failed.
• The author reproduced the failure deterministically and implemented a patch that uses a single Prisma transaction, acquires a PostgreSQL FOR NO KEY UPDATE lock, revalidates state, and persists the entire synthetic batch through the same transaction client.