DataForge Platform
Outcrop connects DataForge to the systems your business runs on
Databases, business applications, files and event streams, on-premises and in the cloud.
Outcrop, how DataForge connects to your systems, has three paths: a connection you configure in DataForge, a connection to the database an application runs on, and the DataForge SDK for any system with an API. The Outcrop agent, one lightweight application that runs inside your network, reaches sources behind your firewall.
Where your data is
Your data center, your cloud accounts, your SaaS vendors
Where it goes
Your Databricks or Snowflake account, and back out to your systems
Three ways to connect
Every connection managed in one place
You configure each connection in DataForge, with its own credentials. Every source feeds governed tables in your account, and every output reads from them. Pick the path that fits the system.
-
01 Configured
Pick it from the list
Choose the system from the connection list, enter the credentials, and schedule the pull. The list covers databases, file storage, Salesforce and the applications built on it, Kafka on Databricks, and your Databricks or Snowflake lakehouse.
-
02 Database
Point a connection at the database the application runs on
An application installed on your own servers keeps its data in a database your team already administers. DataForge reads it with a Table connection and the bundled driver, or with a Generic JDBC connection and the vendor’s driver for any other engine. It reads the tables, keys and relationships and creates the sources for you.
-
03 API
Build it with the DataForge SDK
For a SaaS application or an internal system, your team or ours writes a short notebook in your Databricks or Snowflake account that calls the API and returns a DataFrame. The SDK handles the process lifecycle, logging, credentials and the incremental watermark.
Systems
What DataForge connects to
Here are the common ones. Search the full catalogue for every system and how DataForge reaches it.
-
Microsoft SQL Server -
Oracle Database -
PostgreSQL -
MySQL -
Snowflake -
Databricks Unity Catalog -
Google BigQuery -
Amazon S3 -
Azure Data Lake Storage Gen2 and Blob Storage -
Apache Kafka -
Salesforce - SFTP
-
Microsoft Dynamics 365 and Dataverse -
Microsoft Dynamics GP -
SAP S/4HANA -
SAP ECC -
Oracle NetSuite (SuiteAnalytics Connect) -
Oracle E-Business Suite -
Oracle JD Edwards EnterpriseOne -
Epicor Kinetic (Epicor ERP) -
Infor M3 -
Sage Intacct -
QuickBooks Online -
Workday -
ServiceNow -
Jira -
Zendesk -
HubSpot -
Shopify -
Stripe -
Slack -
SharePoint Online -
Microsoft Excel workbooks -
Google Analytics 4 -
Meta Ads (Facebook and Instagram) -
Google Ads -
Applied Epic -
Guidewire PolicyCenter -
Epic Clarity -
Oracle Health (Cerner) Millennium -
MEDITECH Data Repository -
Yardi Voyager -
Viewpoint Vista (Trimble) -
Procore -
Tyler Munis - Ellucian Banner
-
Blackbaud Raiser's Edge 7 -
PowerSchool SIS
Sources behind your firewall
One lightweight application inside your network
Some of the most valuable data in a company is still stored on a server that security policy keeps off the internet: the enterprise resource planning (ERP) system that runs the plant, the order system that predates the cloud. The Outcrop agent runs on a Windows virtual machine inside that network, or as a managed container in your cloud account.
It reads the databases and file shares there and writes to the storage in your Databricks or Snowflake account. Every connection is outbound over port 443. The agent's configuration file is encrypted and bound to the host machine. Source connections default to read-only intent. Credentials are stored encrypted server-side and fetched each time they are used. You manage the connection and the pipeline that uses it together, in DataForge. The DataForge UI shows the agent as Agent today.
Installed once
It comes as a normal Windows installer, usually for a Windows virtual machine inside your network, or as a managed container in your cloud account. It runs as one application in one process.
One outbound port: 443
The agent opens every connection, and DataForge replies to its heartbeat with commands. There are no inbound firewall rules to open or review.
Updated by the platform
When DataForge ships a release, an agent installed with the Windows installer updates itself to the new version and restarts.
Streaming replication
SQL Server changes, within seconds
For SQL Server, the agent streams changes instead of pulling on a schedule.
It uses log-based change data capture (CDC) and Change Tracking, performs a full initial load, and then continuously merges inserts, updates and deletes into the governed table for that source in your Databricks account. Changes arrive within seconds.
Two tracking methods
Change Tracking is the default. It needs a primary key and snapshot isolation on the source. Log-based CDC needs full CDC enablement. In a locked-down production database, Change Tracking asks much less of your database administrators.
Log-based CDC polls the SQL Server CDC change tables and detects key columns and refresh type from the source configuration. If a retention window expires, the stream reinitializes itself, and a column added upstream is handled automatically.
What streaming replication covers
- Streaming replication covers SQL Server sources. Other database engines are ingested through batch JDBC extraction.
- The streaming target is Databricks. Snowflake deployments use batch ingestion.
- Tables with keys get row-level replication. Keyless tables fall back to full reloads.
Read and write
Reads from any location, writes back to any location
Outcrop is the way in and the way out. It reads from your systems into the Alloy architecture in your Databricks or Snowflake account, where DataForge rules govern every table, and writes the results back out to the databases, files, topics and applications your teams work in.
With other tools, one product ingests, a second transforms, and a third does reverse ETL.
DataForge does all three in one product, and Outcrop is the part that moves the data in and out.
Governed tables
Every source is written to governed tables in your own Databricks or Snowflake account, where your rules, lineage and outputs already work.
Tables
On Databricks, outputs write tables to Delta Lake, Snowflake, SQL Server, PostgreSQL and Google BigQuery. On Snowflake, outputs write Snowflake tables. On both, outputs create views over the governed tables for the BI tools that query in place.
Files
Outputs write files to Amazon S3 or Azure storage for the systems and partners that take them. CSV, Parquet and JSON work on both platforms, and Avro and text files on Databricks.
Events and applications
Outputs publish events to Kafka topics on Databricks, and a Custom Post Output notebook built with the DataForge SDK writes back to any application with an API.
As of July 2026, based on public vendor documentation
In DataForge, every transformation rule writes exactly one column and is validated at save time in one transactional metastore. Your rules read replicated data directly.
DataForge is the only column-level declarative transformation platform with replication built in: a single outbound-only agent inside your network streams log-based SQL Server change data capture straight into governed tables in your own Databricks account, where DataForge rules pick it up. One product, one lightweight agent, no per-row fees.
Data ingested from your sources is stored and processed in your own Databricks or Snowflake account, on AWS, Azure or Google Cloud. DataForge orchestrates the pipelines that run there, on your compute.
Outcrop ships with DataForge. Connections, the agent, the SDK and streaming replication are priced with the platform, and DataForge does not meter rows. If a system needs a third-party driver, you license it from its vendor.
See DataForge pricing →
Bring the systems you run
Tell us the systems your business runs on and where they are, and we will show you how DataForge connects to each one.
Direct answers about Outcrop
What is Outcrop?
Outcrop is how DataForge connects to your systems. It includes the connections you configure in DataForge, database connections for applications that run on SQL Server, Oracle, PostgreSQL, MySQL and other engines, the DataForge SDK for any system with an API, and the agent that runs inside your network to reach sources behind a firewall. Outcrop comes with the DataForge platform.
Which systems can DataForge connect to?
Databases, data warehouses, file storage, event streams, and the business applications that run on them: ERP, CRM, accounting, HR, insurance, healthcare, construction and retail systems. The common ones are on this page. If you can reach a system’s database with a JDBC driver, DataForge reads it with a database connection. If the system has a documented API, DataForge reads it with the DataForge SDK.
My system is not on this page. Can DataForge connect to it?
If it stores its data in a database with a JDBC driver, a Generic JDBC connection reaches it. If it has a documented API, a notebook built with the DataForge SDK reads it and hands the rows to DataForge. Tell us the system and we will tell you which path fits.
Does DataForge connect to applications like Dynamics or Applied Epic?
Yes. Dynamics GP, NAV, AX and Business Central installed on your own servers run on SQL Server, and so does Applied Epic. A Table connection with the bundled SQL Server driver reads them.
Why does the DataForge interface say Agent?
The DataForge UI calls the Outcrop agent Agent today. They are the same thing.
How does DataForge reach a database behind our firewall?
Through the Outcrop agent, one lightweight application that runs inside your network. You install it with a normal Windows installer, usually on a Windows virtual machine. It reads the databases and file shares in your network and writes the data to the storage in your cloud account. Every connection it makes is outbound over port 443, to DataForge and to your storage. There are no inbound firewall rules to open.
Does DataForge do reverse ETL?
Yes. Every DataForge output writes results from the governed tables back to the systems your teams work in, which is what other tools sell as reverse ETL. On Databricks, outputs write tables to Delta Lake, Snowflake, SQL Server, PostgreSQL and Google BigQuery, publish events to Kafka topics, and write CSV, Parquet, JSON, Avro and text files to Amazon S3 or Azure storage. On Snowflake, outputs write Snowflake tables and CSV, Parquet and JSON files. On both, outputs create views in your lakehouse, and a Custom Post Output notebook built with the DataForge SDK writes to any application with an API.
Which databases can stream changes?
Streaming replication covers SQL Server sources, using log-based CDC and Change Tracking. Other database engines are ingested through batch JDBC extraction.
Does streaming replication work with Snowflake?
The streaming target is Databricks. Snowflake deployments use batch ingestion.
Is Outcrop priced separately?
No. Outcrop is part of the DataForge platform and is priced with it. DataForge replication has no per-row fees.
Solution guides
Evaluate DataForge by platform goal
Enterprise data platform
Enterprise data platform for governed analytics at scale
DataForge helps CDOs, CFOs, and data platform leaders scale analytics on one platform in place of separate ETL, orchestration, observability, lineage, and cost-control tools.
Data pipeline platform
Data pipeline platform for complex enterprise source systems
Build, extend, orchestrate, and observe enterprise data pipelines through one consistent architecture while processing data in your Databricks or Snowflake account.
Data engineering platform
Data engineering platform with architecture built in
DataForge is a data engineering platform with automatically enforced architecture, replacing separate ETL, orchestration, observability, catalog, and infrastructure tools while processing data in your cloud.
Data orchestration platform
Data orchestration platform that replaces manually assembled DAG sprawl
DataForge orchestrates data pipelines from structured pipeline definitions, dependency metadata, scheduling, and execution history instead of manually maintained DAGs.
Data observability platform
Data observability platform with lineage, quality, audit, and cost context
DataForge observability ties code, orchestration, quality rules, alerts, lineage, audit trails, and cloud cost visibility back to the platform metadata.
Agentic data engineering
Agentic data engineering platform with save-time validation for AI agents
DataForge lets AI coding agents such as Claude Code and Codex build and operate governed data pipelines through an MCP server that validates every write at save.
Data replication
Data replication that ends inside the governed platform
DataForge replicates SQL Server change data through the Outcrop agent that runs inside your network, and merges it directly into governed tables that the transformation layer already consumes.