Question 1

Does Slingshot look at my company data?

Accepted Answer

No. Slingshot operates entirely off of Snowflake metadata and is careful to never read your actual data. Insights related to query text will only ever use a redacted version of your query with literal values removed.

Question 2

Does Slingshot use my company data to train AI models?

Accepted Answer

No. Slingshot uses commercially available models that are never trained on your data or metadata and do not persist context about your data across sessions. Slingshot uses Snowflake metadata, like table definitions and execution plans, so your actual data is never processed. It never accesses your raw table data or actual query text. Query parameters are redacted at the source.

Question 3

Does Slingshot automatically apply cost-saving changes?

Accepted Answer

Slingshot does not apply changes automatically to your warehouses, but it gives users the option to turn on auto apply for cost-saving recommendations for selected warehouses.

Question 4

How are contextual AI insights different from general SQL AI assistants or LLMs?

Accepted Answer

Basic tools evaluate SQL syntax in isolation. Slingshot provides rich context like execution history and table DDLs to a team of agents that create more comprehensive insights than a simple prompt would.

Question 5

Can Slingshot help me see/understand my AI/Cortex spend/usage/costs?

Accepted Answer

Yes, Slingshot provides deep visibility into the usage of Snowflake Cortex and any agents you've built on Cortex. See which users are consuming which models, understand the cost of your agents and get organizational wide insights into Snowflake Cortex AI usage.

Question 6

Do I maintain control of warehouses that Slingshot is managing?

Accepted Answer

Yes. Slingshot will recommend a schedule for your warehouses, and you can approve the schedule as-is or make any alterations before it is applied. Once a schedule is applied, Slingshot will make changes to your warehouse automatically based on the schedule. You can modify or pause that schedule at any time to regain full control of your warehouse.

Question 7

Can the warehouse recommendations impact performance of the jobs running on them?

Accepted Answer

Slingshot provides smart recommendations with minimal projected impact to the analyzed workloads. You can assess each recommendation's impact on a warehouse before applying one, monitor outcomes after applying them and you can set up an alert to get notified about any SLA breaches for queries on that warehouse.

Question 8

Can Slingshot reduce Snowflake costs without hurting performance?

Accepted Answer

Yes, our recommendations can reduce costs while improving performance. If a recommendation is expected to create a tradeoff between cost & performance, Slingshot will show you the estimated cost and performance impact so that you can make an informed decision for whether to apply the recommendation.

Question 9

How does Slingshot attribute Snowflake spend to teams or business units?

Accepted Answer

Slingshot uses customizable Snowflake tags to map costs directly to your teams, projects or departments. By tracking tagged resources and query activity, you can cleanly allocate shared warehouse spend for showbacks and chargebacks.

Question 10

Can Slingshot show cost and usage across all my Snowflake accounts?

Accepted Answer

Yes, Slingshot gives you a single pane of glass across all your Snowflake accounts and does not require any account or console switching. Our detail pages provide rich insights and metrics into all your resources, and Data Explorer allows you to dig deep into your Snowflake usage.

Question 11

How quickly does Slingshot generate ROI?

Accepted Answer

Slingshot users receive warehouse recommendations within 48 hours of onboarding. Inventory and catalog features and capabilities, like Data Explorer, are available shortly after connecting accounts as well.

Question 12

Does Slingshot provide before-and-after cost analysis?

Accepted Answer

Yes, Slingshot can show you the impact of any change to your warehouse by showing the before and after cost, run-time and normalized cost metrics.

Question 13

How are warehouse schedule recommendations formulated?

Accepted Answer

Slingshot analyzes the workload and query attributes of a warehouse. It tracks factors such as, but not limited to, query size, query load, queued queries and data spillage. Based on these usage patterns, Slingshot generates recommendations that will optimize the Snowflake resources to be dynamically resized based on your actual usage patterns throughout the week.

Question 14

Do recommendations support warehouses with QAS?

Accepted Answer

Yes, we analyze warehouses that have Query Acceleration Services (QAS) enabled and provide recommendations. Additionally, in some cases, we can enhance QAS cost-efficiency by adjusting the time slots in the scheduling recommendations to account for delayed queries or queries that don't meet the eligibility criteria.

Snowflake cost optimization, query tuning and AI workflows with Capital One Slingshot

Why Security and Data Leaders Choose Slingshot

Track Snowflake compute and AI spend in one place

Snowflake AI cost observability with Slingshot Agent Observe

Trusted by industry-leading companies

Maximize your Snowflake data and AI environment

Proactively rightsize compute and warehouse spend

Govern compute and Snowflake AI spend

Secure at enterprise scale

Power your data cloud with contextual AI

See how Slingshot streamlines Snowflake operations

Trusted by technology leaders

Get up and running with helpful Slingshot tools

Try Capital One Slingshot for free

Slingshot docs

Check out our resource center

Slingshot frequently asked questions

Articles, benchmarks and more to help you operate efficiently in the cloud

Snowflake AI cost observability with Slingshot Agent Observe

Snowflake CoCo CLI agent review

Snowflake health check app

Snowflake TPC-DS benchmark SQL worksheet

Slingshot contextual AI drives system-wide data efficiency

Scaling agent context with knowledge graphs on Snowflake

Book some time with the team

Footnotes

Snowflake cost optimization, query tuning and AI workflows with Capital One Slingshot

Why Security and Data Leaders Choose Slingshot

Track Snowflake compute and AI spend in one place

Snowflake AI cost observability with Slingshot Agent Observe

Trusted by industry-leading companies

Maximize your Snowflake data and AI environment

Proactively rightsize compute and warehouse spend

Govern compute and Snowflake AI spend

Secure at enterprise scale

Power your data cloud with contextual AI

Compute and Snowflake AI cost optimization

Compute and Snowflake AI cost attribution

Contextual root cause analysis

Proactive workload tuning

See how Slingshot streamlines Snowflake operations

Trusted by technology leaders

Get up and running with helpful Slingshot tools

Try Capital One Slingshot for free

Slingshot docs

Check out our resource center

Slingshot frequently asked questions

Does Slingshot look at my company data?

Does Slingshot use my company data to train AI models?

Does Slingshot automatically apply cost-saving changes?

How are contextual AI insights different from general SQL AI assistants or LLMs?

Can Slingshot help me see/understand my AI/Cortex spend/usage/costs?

Articles, benchmarks and more to help you operate efficiently in the cloud

Snowflake AI cost observability with Slingshot Agent Observe

Snowflake CoCo CLI agent review

Snowflake health check app

Snowflake TPC-DS benchmark SQL worksheet

Slingshot contextual AI drives system-wide data efficiency

Scaling agent context with knowledge graphs on Snowflake

Book some time with the team

Footnotes