Data pipelines that stay in sync
Synchronoid is a managed platform for building, running and observing data pipelines. Add sources with configuration, keep every step replay-safe, and know where every single value came from.
Private beta · Python 3.10+ · Hosted in the EU (Frankfurt) and US (Virginia)
Recent runs 142 today
Lineage · companies_current
Trusted by beta teams building on Synchronoid
Define assets in Python
Describe the data you want, not the plumbing. Synchronoid works out the order, the retries and the storage.
- Declare a source once with its connector and rate limit. Every pipeline that reads from it shares the same request budget.
- Mark each asset with its layer:
raw,stagedorpublished. Raw payloads are kept, so you can replay any run. - Merge into current records with
track_changes=Trueand get a change history and change events for free. - Run it locally with
syn dev, ship it withsyn deploy.
import synchronoid as sy companies = sy.Source( name="company_registry", connector="http_json", base_url="https://registry.example.org/api", rate_limit=sy.RateLimit(per_minute=120), ) @sy.asset(layer="staged", source=companies, retry=sy.Retry(max_attempts=5)) def company_profiles(ctx, raw): for item in raw.records(): yield { "org_id": item["id"], "name": item["name"].strip(), "address": item["address"], } @sy.asset(layer="published", deps=[company_profiles]) def companies_current(ctx, company_profiles): return company_profiles.merge(key="org_id", track_changes=True) nightly = sy.Schedule(cron="0 2 * * *", target=companies_current)
Built for data that has to be right
Six ideas run through every part of Synchronoid. Together they make pipelines you can trust, explain and leave alone.
Sources by configuration
Add a source with a short YAML or Python config and a small connector. The core never changes, and one connector type can serve many accounts.
Connectors →Provenance on every value
Each field in each record knows its source, fetch time and run id. Ask why a value looks the way it does and get a real answer.
Lineage & provenance →Three storage layers
raw keeps untouched payloads for replay, staged holds typed facts, and published serves merged, current records with history.
Replay-safe steps
Every step is idempotent, so running it twice never duplicates data. If a worker dies, the run resumes from the last finished step.
Orchestration →Rate-limit aware
Connectors share one request budget per source. Add more pipelines and you still never go over a provider’s quota.
Rate limits →Change feeds
Synchronoid compares snapshots and emits business-level events such as “company address changed” to webhooks, Kafka or a pull API.
Change feeds →Ask any value where it came from
Click a field in the control room and see its full history: which source supplied it, when it was fetched, which run wrote it and what it replaced.
- Field-level provenance is stored with the data, not in a separate log you have to join.
- Query it through the API: every published record comes with its provenance attached.
- Trace a bad value back to the exact raw payload and replay just that run.
- Give auditors and downstream teams answers without opening a ticket.
org_id ORG-20418 · address
"14 Harbour Street, Leeds LS1 4AB"
From source to subscriber in three steps
Connect sources
Point Synchronoid at databases, warehouses, files and HTTP APIs. Set a rate limit and a schedule. Credentials live in managed secrets, never in code.
Browse integrations →Model assets
Write Python assets that turn raw payloads into staged facts and published records. Add checks that stop bad data before it reaches anyone.
Follow the tutorial →Publish & subscribe
Downstream apps query the published layer through the API, or subscribe to a change feed and react the moment something changes.
Set up change feeds →Teams that stopped babysitting pipelines
“We pull from many public registries with very different quotas. Shared rate limits meant we finally stopped getting blocked at 3 a.m., and provenance lets our customers see exactly where each record came from.”
Kestrel Property Data
“Reconciliation has to run twice sometimes. With replay-safe steps that is a non-event. We replaced a folder of cron scripts in three weeks.”
Northwind Ledger
“Meter data arrives late and out of order. Durable runs resume where they stopped, and the change feed tells our grid team only what actually changed.”
Helio Grid Analytics
Works with the stack you already run
Read from databases, warehouses, object storage and APIs. Write to your warehouse, stream changes to Kafka or webhooks, and send alerts where your team already looks.
Questions teams ask us
Who can join the private beta?
Teams that run production data pipelines and want to test Synchronoid on real sources. We open new workspaces in small batches every few weeks. Join the waitlist and tell us about your data.
Do I need to rewrite my existing pipelines?
No. Most teams start with one or two sources that cause the most pain and keep the rest as they are. Synchronoid works next to dbt, your warehouse and existing schedulers, and you can import Airbyte connections.
Where is my data stored?
You choose a region when you create a workspace: EU (Frankfurt) or US (Virginia). Data stays in that region. We sign a data processing agreement with every customer.
How is Synchronoid different from Airflow?
Airflow schedules tasks. Synchronoid manages data assets: it stores raw payloads, tracks provenance per field, shares rate limits across pipelines and emits change events. See our comparison for details.
What does it cost during the beta?
The Explorer plan is free during the beta. The Team plan is $490 per month, which is 30% below the list price. Enterprise pricing is custom. See pricing.
Is Synchronoid SOC 2 certified?
Not yet. Our SOC 2 Type II audit window opened in Q3 2026 and is in progress. We are GDPR-ready and publish our security practices on the security page.
Latest from the Synchronoid Team
Synchronoid v0.14: change feeds, cost insights and new connectors
Kafka and webhook change feeds, a cost insights preview, four new connectors and a faster CLI.
Rate limits are a feature
Why we treat a provider’s quota as a shared budget, and what that changes for teams with many pipelines.
Provenance for every value
How we store source, fetch time and run id next to each field without slowing down your queries.
Ready to stop babysitting pipelines?
Synchronoid is in private beta. Tell us about your data and we will get you a workspace.