01 — Overview
Your warehouse probably holds the best version of the truth, and most of the company sees it only as a dashboard. Chamfer turns Snowflake tables into working tools: a claims queue ordered by risk score, a pricing review with margin by SKU, an accruals sheet that finance can correct and approve. Each tool reads the same modeled tables your analysts already maintain, so the numbers match the dashboards people know.
Chamfer queries Snowflake with the role and virtual warehouse you assign, so existing grants and masking policies still apply, and credit usage appears on your own bill. Set a row limit and a query timeout per app to keep costs predictable. When an app needs to write, such as a corrected forecast or an approval flag, it writes only to the schemas you allow, and every write lands in the Logbook.
02 — What you can do
Your role, your warehouse
Queries use the role and warehouse you choose, so grants, masking and credits stay yours.
Cost limits per app
Set row limits, timeouts and caching so a busy screen never runs up a surprise bill.
Controlled write-back
Apps write corrections or approvals to named schemas only, each write logged by user.
03 — FAQ
Before you connect Snowflake
Which authentication methods work with Snowflake?
Will Chamfer apps run up our Snowflake bill?
Do our masking policies still apply?
Can apps write back to Snowflake?
04 — Pairs well with
Teams often connect these next
05 — Used in
Dashboards on warehouse data that people can act on, with access by team, region or site written as one rule.
Route refunds, discounts, vendor changes and releases to a named approver, with the diff and the decision on record.
Turn the shared sheet everyone is afraid to touch into an app with validation, roles and a named owner.
Put Claude or GPT to work on your own data, and let Operators work a queue within the tables and actions you allow.