Misata as a Neosync Alternative
Neosync's open-source repo was archived on August 30, 2025 after the team joined Grow Therapy, and its hosted cloud went offline. If you relied on it, you need a new home for two different jobs it did: generating synthetic data for dev and staging, and anonymizing production data. Misata covers the first job well and differently: it generates whole relational datasets from a schema with no production data involved at all, guarantees foreign-key integrity with a proof you can assert in CI, and seeds Postgres, MySQL, or SQLite directly. For the second job, masking real production rows, Misata is honestly not the tool; pair it with a masking tool like Greenmask if you need that half.
| Capability | Misata | Neosync |
|---|---|---|
| Actively maintained | Yes, frequent releases | No, repo archived Aug 2025 |
| Generate synthetic relational data from scratch | Yes, schema or plain English | Yes, per-column generators |
| Needs production data access | No, nothing to connect or leak | Optional, for anonymize/sync |
| Foreign-key integrity | Guaranteed, orphan counts reported | Supported via subsetting |
| Declared outcomes (rates, revenue curves) | Yes, exact by construction | No |
| Seed a database directly | Postgres, MySQL, SQLite | Postgres, MySQL, S3 |
| Anonymize and sync production data | No, use a masking tool | Yes, its core strength |
| Deterministic and seeded | Yes, identical bytes per seed | Partial |
What happened to Neosync
Neosync built a well-liked open-source platform for anonymizing production data, generating synthetic data, and syncing environments. In September 2025 the team was acquired by Grow Therapy; the GitHub repository was archived on August 30, 2025 and the hosted service shut down. Archived code still runs, but there are no more releases, security patches, or fixes, which is a hard position for a tool that sits in your data path.
Where Misata covers the gap, and where it does not
If your Neosync use case was seeding dev and staging with realistic, referentially intact data, Misata does that job without ever touching production: you describe the schema (or point it at your database's structure) and a deterministic engine generates the rows, with an integrity report per relationship. If your use case was masking real production rows in place, that is a different job and Misata does not pretend to do it; Greenmask is the actively maintained open-source option for pg_dump-compatible masking. Many teams end up with both: masked production for pre-release checks, generated data for everything earlier in the pipeline.
The part Neosync never had
Misata generates data backwards from a declared outcome. A 5 percent return rate comes out at exactly 5 percent; monthly revenue that should rise from 80k to 450k lands on those numbers to the cent. That turns seed data into a known-answer test: run your pipeline over data whose correct answer you decided in advance, and assert the output matches. Neosync seeded environments; Misata also seeds assertions.
Frequently asked
Is Neosync still maintained?
No. The repository was archived on August 30, 2025 after the team was acquired by Grow Therapy, and the hosted cloud service is offline. The code remains public but receives no updates.
Can Misata anonymize my production database like Neosync did?
No, and it does not try to. Misata generates data from a specification, so no production data is needed or touched. For masking real production rows, pair it with a dedicated masking tool such as Greenmask.
Can Misata seed my Postgres dev database directly?
Yes. misata.seed_database(schema, "postgresql://...") inserts generated tables in dependency order so foreign-key constraints hold, and the same seed reproduces the same database every time.
Compare Misata with