Saddle Data Pipelines

Saddle Data Pipelines was our first product: a zero-trust data integration platform for SREs and data teams. We have set it aside so we can focus on Ultraviolet Observability, and new sign-ups are closed.

Still interested? If Saddle Data Pipelines would solve a real problem for your team, get in touch. We’re open to bringing it back for the right use case.

What it does

  • Secure Remote Agents: A single Go binary runs inside your VPC and extracts data over outbound-only connections, so you never open an inbound firewall port.
  • Automated PII Enforcement: Tag a column as sensitive once, and masking or hashing is applied in every flow that uses it.
  • Global Impact Analysis: A visual dependency graph shows which downstream assets will break before you change a source table.
  • Schema Drift Tracking: An automated, human-readable audit log of when and how upstream schemas changed.
  • In-Flight AI Embeddings: Turn raw text into vector embeddings mid-stream for RAG applications.
  • Broad Connector Library: PostgreSQL, MySQL, MongoDB, Snowflake, BigQuery, Databricks, ClickHouse, Salesforce, HubSpot, vector databases, and more.

Read more

Our blog has deep dives on the architecture and features behind Saddle Data Pipelines.

Contact us about Saddle Data Pipelines →