Data lake and analytics · AutessaDB
Semantic layer
The AutessaDB semantic layer teaches people and AI agents how to query the data lake correctly: a description and grain, column descriptions and synonyms, named measures, documented joins and verified queries. Suggest with AI drafts it from the table's columns, statistics and a sample, and nothing is saved until you review it.
-- the problem
Text-to-SQL is only as good as what the model knows
Ask an agent for "net sales last quarter" and it has to guess which column means net sales and how to total it. A semantic layer answers those questions once, in your own business terms, and agents prefer a verified query when a question matches one, so answers stay consistent.
What you get
- Description and grain: what one row represents, such as one row per order line.
- Column descriptions and synonyms for the words people actually use.
- Measures such as revenue := SUM(net_amount), offered as completions and listed for agents.
- Joins, such as order_lines.customer_id = customers.id, many-to-one.
- Verified queries pair a plain-language question with checked SQL, and are re-validated when the schema changes.
- Suggest with AI drafts each field next to the current value with a checkbox; Apply selected fills the form.
- Suggestions only fill empty fields by default and do not count toward AI usage.
- From the CLI: autessa analytics suggest, with --focus measures or --apply.
-- how it works
How does semantic layer work in AutessaDB?
Open the layer
On the table, open Edit, then Semantic layer.
Draft with AI
Click Suggest with AI and review each suggestion; Test verified queries.
Apply and save
Apply selected, adjust, then Save semantic layer.
Put it to work
Ask, the workbench and data agents all read the layer.
-- every option
Semantic layer fields
| option | what it does |
|---|---|
| Description / grain | What the table holds and what one row is. |
| Column descriptions and synonyms | Meaning and other names for each column. |
| Measures | Named SQL aggregate expressions. |
| Joins | How this table relates to others, with cardinality. |
| Verified queries | Question plus checked SQL; agents prefer them. |
-- example
Full reference in the AutessaDB CLI docs and the API reference.
-- permissions
Who can do what?
| access | can |
|---|---|
| ADMIN | Edit and save the semantic layer. |
| READ | See it in describe results and completions. |
◆ AI assist
Where AI helps
Suggest with AI reads the table's columns, statistics and a 50-row sample, as you are allowed to see them, and drafts every field for review. Ask and agents then use the layer to write SQL.
AI in AutessaDB works inside the same access rules as the person using it, and schema changes always wait for a person to apply them.
-- faq
Questions about semantic layer
What is a semantic layer?
Metadata that explains a table: what a row is, what columns mean, named measures, joins and verified example queries, so people and agents query it consistently.
Does Suggest with AI change my table?
No. Suggestions are drafts with checkboxes; nothing is saved until you click Save semantic layer.
Do agents use verified queries?
Yes. Agents prefer a verified query when a question matches one.
100% customizable, code and all
Make AutessaDB fully your own.
Use it out of the box, then change anything. Ask Autobot, Autessa’s AI builder, for a new field, screen, report, workflow or agent and it builds it into your copy of the app. Or open the code and edit it yourself. Nothing is locked: the app, its data and its code are yours.
Start free and make it yours