Why Monitoring Data Goes Unused.

How Unleash Live and Snowflake Fixed the problem.

A resources sector operator running a fleet of drilling rigs had sensors installed and telemetry running across every rig. Almost nobody logged in to check it.

The readings were real: water pressure, rotation speed, feed rate, captured and converted from analog dials to digital data. But that data lived in a separate monitoring portal, one more browser tab someone had to remember to open. Getting value out of it meant exporting it by hand and lining it up against other systems, an extra step with no link into how the team actually worked.

Automating that flow directly into the cloud-based data platform, Snowflake, a system the team already used every day, removed the extra tab and the manual export step entirely.

Why Snowflake, and why now

Snowflake is where many enterprises already centralize data from across the business: ERP, historian, SCADA, and everything in between. If a team already trusts Snowflake as its system of record, adding another siloed dashboard on top of it works against the very fragmentation most operators are trying to avoid.

Prism is Unleash Live's always-on monitoring product: fixed cameras, continuous visual coverage, turning what a camera sees into a structured reading the moment it happens. On its own, that reading lands in Unleash Live's dashboards and alerts, useful, but still one more system to remember to check.

Feeding Prism's readings directly into Snowflake removes that extra step. The computer vision data appears in the same place where a team already runs every other query.

What Prism delivered

For this operator, Unleash Live deployed Prism for automated analog gauge monitoring: reading and digitizing pressure and equipment gauges that would otherwise need a manual check on site.

Unleash Live covered a related use case in an earlier blog on: automated industrial meter reading.)

That gauge data now feeds directly into Snowflake, giving the operator a single, queryable home for readings that used to sit siloed out in the field.

Before expanding coverage, adding another camera for continuous gate monitoring, for example, the operator wanted to benchmark these automated readings against what they already had. Proving the value first kept the rollout low risk.

What changed for the team.

The immediate win is structural: readings that used to require a manual pull now arrive automatically in Snowflake, in the same place the team already runs its other analysis.

Unleash Live is also working with Snowflake to extend this toward natural language querying, so the same data could eventually be asked about directly rather than pulled into a custom report each time. That would mean questions like "Were there any anomalies this week?" or "What was the peak pump pressure?", getting answered in plain language.

Every reading can still be traced back to the source footage. If a pressure spike appears in the data, the underlying clip, either the AI-annotated version or the raw feed, is available to check. The visual evidence stays attached to the number.

How the integration works

Unleash Live's existing Analytics API is the connector, so bringing gauge data into Snowflake required no major new infrastructure. Data syncs on a scheduled cadence via a Snowflake task, keeping the dataset current without anyone needing to run a manual export.

The same pattern is portable. Customers on Databricks can move gold layer data into Snowflake via a documented path, and a parallel path exists for Redshift, so the approach isn't tied to a single warehouse. Visit our MCP deployment and our integration page for dozens of further connectors or request your custom connector.

Measured in the field:

• Value of portfolio monitored: $69B
• Reduction in manual inspection effort: 70%
• Images detected and analyzed: 97B
• Detected F1-score across deployed AI Apps: 91%

Ready to see Unleash Live in action? Request a technical briefing.