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: SQL
Metadata Filtering for Databricks RAG Using ai_extract and ai_prep_search
Combine ai_parse_document, ai_extract v2.1 and ai_prep_search so each chunk carries typed metadata, letting Vector Search filter before ranking by similarity.
7 min readClickHouse MCP and the Limits of Read-Only: Structured Access for Agents
Read-only MCP access stops writes, not wrong numbers, tenant leaks or schema exposure. See what the ClickHouse benchmark shows and how structured metric tools reduce the risk.
8 min readApache Doris 4.1: Vector Search, JSON Storage, OLAP and Lakehouse
A walkthrough of what Apache Doris 4.1 adds: IVF vector indexes, BM25 search(), 100 MB JSON documents, Segment V3, query caches, Iceberg writes, MERGE INTO and TIMESTAMPTZ.
18 min readRAG Without a Vector Database: A Pipeline in BigQuery SQL and Where It Breaks
Build retrieval-augmented generation entirely in BigQuery SQL, from embeddings to grounded answers, then see the costs, quotas and prompt-versioning limits.
11 min readPostgreSQL MCP in Practice: Architecture, Security, Best
PostgreSQL MCP in practice: stop giving AI raw database access, what makes the server different, where it fits next to a REST API, security and best practices.
15 min readPostgreSQL + pgvector and SQL Server 2025 as Vector Stores
PostgreSQL 18 with pgvector vs SQL Server 2025 as vector stores: how each works as a vector database, installing pgvector on Windows, and storing embeddings.
18 min readPractical notes: Production-Grade Text-to-SQL Agent with Claude Code
Operable walkthrough of Practical notes: Production-Grade Text-to-SQL Agent with Claude Code: contracts, checks, and drop-in code slots for teams shipping this pattern.
18 min readPractical notes: Text-to-SQL Agent in Python: LLM Tool Calling Tutorial
Operable walkthrough of Practical notes: Text-to-SQL Agent in Python: LLM Tool Calling Tutorial: contracts, checks, and drop-in code slots for teams shipping this pattern.
13 min readAfter the LangGraph Tutorial: Hard Edges, Schemas, and Safety Layers
Taking a LangGraph SQL agent past the tutorial: typed guards, structured output as a contract, schema refresh, cost routing and layered checks before prod.
12 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.