---
title: "Reducing Snowflake burn with a context harness for the warehouse | Rig"
description: "A mid-market SaaS firm moved AI exploration off Cortex and onto Rig. Warehouse-wide context and an in-process sandbox cut repeated live queries and gave agents shared business definitions. The implementation’s cost model projects 84% lower AI-serving spend and about 29% lower total Snowflake spend at the same adoption."
canonical: "https://rig.so/case-studies/reducing-snowflake-burn"
format: markdown
---

AI enablement

# Reducing Snowflake burn with a context harness for the warehouse

Rig engineers delivered the implementation with the firm’s IT and Data teams, starting with an audit and proof of concept.

A mid-market SaaS firm moved AI exploration off Cortex and onto Rig. Warehouse-wide context and an in-process sandbox cut repeated live queries and gave agents shared business definitions. The implementation’s cost model projects 84% lower AI-serving spend and about 29% lower total Snowflake spend at the same adoption.

**84%**projected reduction in AI-serving spend

**~29%**projected reduction in total Snowflake spend

**60%+**of tool calls handled without warehouse execution in Rig’s benchmark

Same adoption · modelled Snowflake consumption

## More AI use before the credits run out

### Total spend over the modelled term

Baseline indexed to 100

Cortex / raw MCP**100**

AI through Rig**71**

About 29% lower total spend, with the same base workloads and adoption curve. The scenario projects roughly seven to eight months of additional credit runway.

### Snowflake cost per question

Cortex Agents · audited

~$0.50 token fees, SQL compute extra

Rig · serving estimate

$0.01–$0.07 warehouse serving only

This comparison covers Snowflake consumption only. Rig fees, external model costs and context builds are excluded from the serving estimate.

Billing mechanics: [Snowflake Cortex Agents](https://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents) and [warehouse billing](https://docs.snowflake.com/en/user-guide/warehouses-overview). Unit costs reflect the audit, not current list prices.

[AI adoption was accelerating Snowflake burn](#story-section-0)[An audit with IT and Data established the opportunity](#story-section-1)[Rig became the harness for the data warehouse](#story-section-2)[The in-process context sandbox does the preparation](#story-section-3)[Less Snowflake compute per question](#story-section-4)[A slower credit burndown at the same adoption](#story-section-5)[Shared definitions support more accurate answers](#story-section-6)[Measurement](#story-section-7)

## AI adoption was accelerating Snowflake burn

More people were using AI on the firm’s data, and the credit budget was coming under pressure. AI-related workloads had grown from roughly 4% to about half of consumption in the audited account. Cortex Agents consumed credits for inference, with warehouse compute billed on top. Agents connecting directly through MCP also ran live queries to discover schemas, sample data and retry SQL.

## An audit with IT and Data established the opportunity

Rig started with the IT and Data teams. Together, they reviewed account metering, Cortex request charges and warehouse idle time, then tested the Rig approach in a proof of concept. The audit validated the cost-saving opportunity against observed query costs. Rig built context across the entire warehouse, bringing table relationships, joins and business definitions into one place for agents to use.

## Rig became the harness for the data warehouse

Rig wraps the warehouse with the context, tools and access controls an agent needs to work with company data. The agent can understand the available data, use agreed definitions and check its approach before it sends SQL to Snowflake. The warehouse continues to serve live data. Rig handles the surrounding exploration and preparation, so each new question does not require the agent to rediscover the same relationships through paid queries.

## The in-process context sandbox does the preparation

The sandbox runs inside Rig’s agent process, alongside the warehouse context. Agents can inspect that context, test query logic against available samples and shape results there before executing live SQL. These steps avoid warehouse execution when the context or samples are sufficient. The final query goes to Snowflake through governed tools. A shared serving warehouse also reduces separate cold starts, which matter because Snowflake charges a 60-second minimum each time a warehouse starts.

## Less Snowflake compute per question

The audit put Cortex Agents at about $0.50 per request in token fees, before SQL compute. Rig’s estimated Snowflake serving cost was $0.01 to $0.07 per question, informed by observed account query costs and Rig’s session benchmark. In that multi-tenant benchmark, more than 60% of tool calls handled context or shaping without executing warehouse queries, and 22% of sessions needed no live query. Moving preparation into the sandbox reduces warehouse hits and the compute time spent on exploratory work.

## A slower credit burndown at the same adoption

The cost model kept the base workload and AI adoption curve the same on both paths. Serving AI through Rig projected 84% lower AI-serving spend and about 29% lower total Snowflake spend over the modelled term. The credit pool lasted roughly seven to eight months longer in the scenario. That gives the firm more room to expand AI access before buying additional Snowflake capacity.

## Shared definitions support more accurate answers

The same implementation addressed answer quality. Warehouse-wide context gives agents the joins and business definitions behind a question, and the sandbox lets them test query logic before live execution. IT and Data retain control of definitions and access as more people use AI on the warehouse.

## Measurement

The comparison combines the account audit with Rig’s multi-tenant session benchmark. Savings and credit runway are projections at equal adoption and the audit’s credit price. Per-question estimates cover Snowflake serving consumption; Rig fees, external model costs and context builds are separate. The model uses a hosted Rig benchmark for its AI-serving assumption.

Talk to Rig

## What work do you want to improve?

Tell us where your team spends time, loses money or needs better information.
In a 30-minute call, we’ll discuss the data and systems work that could help.

[Discuss your project](https://rig.so/book-demo)Use the platform with your team, or add our engineers.

---

- [This page as HTML](https://rig.so/case-studies/reducing-snowflake-burn)
- [Site map for language models](https://rig.so/llms.txt)
- [API and agent documentation](https://rig.so/developers)
