Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: Pinecone
Semantic Search Service with a Pinecone Vector Database Behind an MCP Server
A reusable semantic search service: chunk PDFs, store embeddings in a Pinecone vector database, and let any agent query them through a FastMCP server.
7 min readMoving RAG From Pinecone to Postgres pgvector: Six Months Later
Moving RAG from Pinecone to Postgres pgvector, six months on: 9M vectors, real cost, latency and recall numbers, what got worse, and when pgvector is right.
7 min readReranking in RAG with Spring AI and Pinecone: Moving an Answer from Rank 4 to 1
Reranking in RAG with Spring AI and Pinecone: plain vector search vs a bge-reranker step on an HR handbook, where a buried answer moves from rank 4 to 1.
6 min readAgentic RAG in LangGraph: Graders, Conditional Routing and Web Search Fallback
Move a linear LangChain RAG pipeline to a LangGraph StateGraph with LLM graders, conditional entry and edges, tests and a FastAPI endpoint that reports provenance.
7 min readChoosing a Vector Database at Scale: Weaviate vs Pinecone for RAG
A practical comparison of Weaviate and Pinecone for growing RAG systems: hybrid search, tenant isolation, memory trade-offs, deployment options and how to benchmark.
9 min readPinecone, Weaviate, Milvus, Qdrant or Chroma: Picking a Vector Database
Plain-language meaning of each vector database pitch, from serverless and hybrid search to Rust and GPU indexing, with trade-offs and when each one fits best.
8 min readClaude Custom MCPs for production systems
Custom MCP connectors for Claude in production: the three problems they solve, from scattered business info to missing context, and how each server fixes one.
11 min readOne Retriever, Three Backends: Porting NumPy Search to Chroma, FAISS, Pinecone
Port a hand-rolled NumPy cosine retriever to Chroma, FAISS and Pinecone, and learn what indexes, metadata filters and persistence add, plus the mistakes that break relevance.
11 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.