Topic 01 · Autessa Synthetic Data
Describe and design
Autessa Synthetic Data starts from a sentence: describe the data, let AI propose the columns, refine them in the schema editor and steer the style with a few example rows.
In this area
What is in describe and design?
Turn a plain-English description into a synthetic dataset: AI-proposed schemas, a column editor, few-shot sample rows and a training pairs preset for LLM data.
In this topic
- Autessa Synthetic Data generates realistic rows from a plain-English description of the domain, distributions and edge cases you want.
- Autessa Synthetic Data proposes a table name and typed columns from your description, so you can start from a draft schema instead of a blank one.
- The Autessa Synthetic Data schema editor defines each column with a type, a description that guides the generator, a required flag and allowed values.
- Autessa Synthetic Data accepts a JSON array of sample rows as few-shot examples that steer the style and format of every generated row.
- Autessa Synthetic Data generates LLM training data: a one-click preset sets up system, prompt and response columns and exports the pairs as JSONL or CSV.
Subtopics
Every describe and design feature
Generate data from a description
Autessa Synthetic Data generates realistic rows from a plain-English description of the domain, distributions and edge cases you want.
Read more →AI schema proposal
Autessa Synthetic Data proposes a table name and typed columns from your description, so you can start from a draft schema instead of a blank one.
Read more →Schema editor and column types
The Autessa Synthetic Data schema editor defines each column with a type, a description that guides the generator, a required flag and allowed values.
Read more →Sample rows as few-shot examples
Autessa Synthetic Data accepts a JSON array of sample rows as few-shot examples that steer the style and format of every generated row.
Read more →LLM training data and prompt pairs
Autessa Synthetic Data generates LLM training data: a one-click preset sets up system, prompt and response columns and exports the pairs as JSONL or CSV.
Read more →Keep exploring
Other areas of Synthetic Data
Where data lands →
Every Autessa Synthetic Data job has one target: an existing table, a new table it creates first, a downloadable JSONL or CSV file, or a 50-row quick generate that stores nothing.
Quality and variety →
Autessa Synthetic Data checks every generated row against your columns, repairs or drops the ones that do not fit, and removes near-duplicates so a 5,000-row set does not repeat itself.
Review before write →
In review-first mode nothing is written until a person approves: preview the staged rows, exclude the misses, regenerate a batch or append more, then approve or discard.
Jobs at scale →
Autessa Synthetic Data runs each job in batches across several workers, shows live progress, resumes stalled work automatically and always ends in a clear status.
Developers and governance →
Everything in Autessa Synthetic Data is scriptable from the Autessa CLI and REST API, and it runs on the model your company chooses, with optional PII and PHI stripping.
100% customizable, code and all
Make Autessa Synthetic Data 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