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: data-engineering
Agentic Analytics Is More Than Text-to-SQL: Lessons From Anthropic
What Anthropic's self-service analytics post says about semantic layers, skills, evals and data foundations, and why text-to-SQL alone gives a false sense of precision.
17 min readGoogle Next 2026 for Data Engineering: Looker, BigQuery and MCP Servers
What Google Cloud Next 2026 means for data engineering on Looker and BigQuery: agents, the LookML semantic layer, managed MCP servers and the Knowledge Catalog.
14 min readAgents Need Interpretable Meaning: Semantic Layers, Ontologies and Governed Metrics
Conversational BI fails when meaning is unresolved. What text-to-SQL studies show, and how a semantic layer, an ontology and canonical IDs make answers safe.
15 min readSearchable Long-Term Memory for Data Engineering Agents
How a multi-agent data engineering system stopped re-debugging the same failures: automatic capture, scoped diaries, a curated tier and context-level recall.
8 min readStop Your RAG From Serving Stale Documents: Recency and Staleness Tracking
Stop RAG from serving stale documents: change capture, staleness flags and recency-weighted reranking on Databricks Vector Search so agents use current docs.
9 min readTurning Millions of Documents Into an Agentic Knowledge Graph Here’s What I
Operable walkthrough of Turning Millions of Documents Into an Agentic Knowledge Graph Here’s What I: contracts, checks, and drop-in code slots for teams shipping this pattern.
6 min readPractical notes: MCP for Analytics: The Protocol Solves Connectivity, Not
Operable walkthrough of Practical notes: MCP for Analytics: The Protocol Solves Connectivity, Not: contracts, checks, and drop-in code slots for teams shipping this pattern.
8 min readPractical notes: The Semantic Layer, Explained: The Meaning Layer Your AI Agent
Operable walkthrough of Practical notes: The Semantic Layer, Explained: The Meaning Layer Your AI Agent: contracts, checks, and drop-in code slots for teams shipping this pattern.
9 min readSelf-healing schema drift with Airflow, LangGraph, and MCP pull requests
When ETL fails on a column change, detect, classify drift, draft a code fix, open a GitHub PR, and Slack the team—without writing straight to production.
4 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.