Our engine never sees your warehouse password
datahashi

Blog

Notes on governed metrics

How we model, query and secure the numbers that dashboards and AI agents share. From the people building datahashi.

Agents & MCP

Grounding a support copilot in certified customer health metrics

How a support copilot answers an account-health question from certified metrics, with sample schema, model and queries that separate ticket volume from open critical escalations.
Agents & MCP

Self-hosting regulated AI agents without leaking warehouse credentials

How a healthcare compliance team gave LLM agents access to clinical treatment analytics inside a private VPC without exposing database passwords.Governance & trust

How to survive warehouse schema changes without breaking dashboards

Upstream dbt jobs renamed an invoice column to amount_usd, but central semantic mapping and model tests kept dashboards running without a blip.
Modeling & metrics

Why marketing CAC and ROAS never matched finance

How governed derived metrics and transparent lineage unify customer acquisition cost and return on ad spend across marketing channels and finance.Querying

Serve a morning dashboard from rollups, not raw scans

How pre-aggregation rollups serve morning executive dashboard queries from pre-computed tables rather than scanning raw transaction tables.Governance & trust

Multi-tenant customer dashboards: row-level policy isolation

How platform engineers secure embedded customer dashboards using JWT claims, row-level policies, and query metering in a thin semantic layer.Modeling & metrics

E-commerce category tree revenue: safe hierarchical rollups

How retail merchandisers drill down SKU, category, and department revenue without fan-out bugs in a thin semantic layer.Modeling & metrics

SaaS activation and retention without 14 CTEs

How product managers analyze SaaS user activation and weekly retention across entity join graphs in a thin semantic layer.Agents & MCP

Slack bot weekly results: certified metrics for operations

How operations teams connect Slack bots to certified metrics using Model Context Protocol and query time zones in a thin semantic layer.Modeling & metrics

Finance close: one ARR across board decks and dashboards

How semi-additive snapshot measures and model tests keep ARR balances consistent across reporting tools in a thin semantic layer.Warehouses & setup

BigQuery least privilege for agents

Apply BigQuery least privilege so agent-backed semantic serving never needs broad project roles or warehouse passwords in the agent.Warehouses & setup

Snowflake read-only roles for a semantic layer

A semantic layer on Snowflake is only as safe as the role that runs its queries. datahashi's expected posture is least-privilege, read-only roles for query execution.

Topics