Pimcore vs PostgreSQL

UK Investor Tenzing Acquires Majority Stake in Austrian MarTech Pimcore

UK-based technology investor Tenzing has acquired a majority stake in Salzburg-based data management software provider Pimcore. Pimcore, founded in 2009, operates an open-core platform that unifies product, customer, and digital asset data into a single source of truth, serving over 400 enterprise clients globally. The company was previously backed by Munich-based Nordwind Growth, which led a $12 million Series B in 2022 and held approximately a 73% stake. Pimcore's founders, Dietmar Rietsch and Matthias Blauth, will remain on board as co-CEOs to guide international expansion and target potential acquisitions, positioning clean data as a crucial baseline for enterprise AI applications. The acquisition is Tenzing's second deal in the DACH region and is currently pending antitrust regulatory clearance.

• UK tech investor Tenzing is acquiring a majority stake in Salzburg-based Pimcore.
• Previous majority shareholder Nordwind Growth led a $12 million Series B in Pimcore in 2022 and held a 73% stake.
• Pimcore transitioned from an open-source community model to an open-core approach, driving organic growth.

We built the layer enterprise AI is about to crash into. Now we scale it.

Today Nordwind Growth announced an agreement to sell its majority stake in Pimcore to Tenzing, a London-based investor in European B2B software. The transaction remains subject to antitrust approvals.

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.