Skip to content
datarelix.ai docs

Introduction

Welcome to datarelix.ai

datarelix.ai is an AI-powered data client that transforms natural language questions into validated SQL/KQL queries and optional Python analysis.

What is datarelix.ai?

datarelix.ai lets you ask questions about your databases in plain English. The AI understands your question, generates the appropriate query, executes it securely, and optionally performs Python analysis to produce charts, statistics, and insights.

Supported databases

  • PostgreSQL
  • MySQL
  • Azure SQL / Synapse
  • Databricks SQL
  • Kusto (Azure Data Explorer, KQL)
  • BigQuery
  • Amazon Athena
  • Elasticsearch (ES|QL, 8.11+)
  • Snowflake

Key features

  • Natural language queries: ask questions like “What were our top customers last month?”
  • Validated execution: every generated query is parsed and validated before it runs.
  • Sandboxed Python analysis: charts, statistics, and complex transformations in an isolated runtime — no network access, strict resource limits.
  • Multi-dialect: one product, nine query languages and engines.
  • Enterprise auth: Google, Microsoft Entra (Azure AD), and GitHub SSO.
  • API-first: integrate with your applications via REST.
  • Fully managed: datarelix.ai runs the orchestrator, query services, and sandbox — nothing to install or operate.

How it works

  1. Connect — add your database credentials (encrypted at rest).
  2. Ask — type a question in natural language.
  3. Plan — the LLM produces a structured plan (Intermediate Representation), not direct tool calls.
  4. Execute — the orchestrator validates and runs each step against the database (via MCP) or in the sandbox (for Python).
  5. Answer — you get an artifact-first result: tables, charts, and a short explanation, with full provenance.

Architecture at a glance

datarelix.ai is built around a strict execution boundary: the LLM never executes tools directly.

User question
    ↓
Orchestrator (FastAPI)
    ↓
LLM produces a structured plan (IR)
    ↓
Orchestrator validates and executes each step
    ├─→ MCP servers (per-dialect SQL execution)
    └─→ Sandbox (isolated Python runtime)
    ↓
Artifact-first result

This separation means:

  • The LLM never has database credentials.
  • All SQL is validated by sqlglot’s AST parser before execution.
  • Python runs in a sandbox with no network access and strict CPU/memory limits.
  • Every run is fully inspectable as a step-by-step trace.

Read more in Core Concepts.

Getting started

Ready to start? Head to the Quick Start Guide to set up datarelix.ai in minutes.