INVOICE PROCESSING

Suggest GL codes from a vendor's historical coding in Snowflake

For each new invoice line, queries Snowflake for how that vendor and similar line descriptions were coded before.

CategoryInvoice Processing
Enginesim
Difficultyadvanced
Triggerwebhook
Steps5
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerNew invoice received via webhookHTTP webhook
  • ActionQuery Snowflake for vendor coding historySnowflakeSnowflake
  • ActionReconcile history with current line textOpenAI
  • LogicSplit into auto-suggest vs needs-review
  • OutputWrite recommended codes to AirtableAirtableAirtable

What it does

Gives each invoice line a GL-code suggestion grounded in your own ledger history. Rather than guessing from text alone, it asks Snowflake how the same vendor and similar descriptions were coded in prior periods and recommends the dominant account.

When to use it

Use it when you have years of coded invoice history in a warehouse and want suggestions that match how your team actually codes, not generic guesses. Ideal for high-volume recurring vendors.

How it works

  1. 1A new invoice record (via webhook from your intake system) triggers the run.
  2. 2For each line, a Snowflake query pulls historical GL codings for the same vendor and fuzzy-matched line descriptions.
  3. 3An OpenAI step reconciles the history with the current line text, picking the most consistent account and explaining the choice.
  4. 4A logic step splits lines into auto-suggest (clear historical match) and needs-review (new or ambiguous).
  5. 5The recommended codes and rationale are written back to Airtable for the AP team to confirm.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect HTTP webhookTrigger any URL on agent actions.
  2. 2
    Connect SnowflakeWarehouses, queries, shares.
  3. 3
    Connect OpenAIModels, embeddings, files.
  4. 4
    Connect AirtableBases, tables, views, automations.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.