An example setup for a finance team that compares bank statements with the books by hand every month: the files are yours, the list of differences is on your desk in the morning.
It runs on your own server: work files and records are stored on your infrastructure; content processed by AI goes to the model provider you choose, and data sent to external tools goes to the connections you set up.
Chat · finance-ops
demo data
I uploaded the October statement and the ledger export. List the unmatched entries with possible reasons.
16:10
Remyral16:10
I drafted a job in the finance project, which has the network off. Approve it and I’ll start.
Job draftAwaiting approval
October reconciliation
Project
finance-ops · network off
Input
statement-oct.csv · ledger-oct.csv
Step
Python · read-only
Output
differences.csv
Job result · 16:24Completed
13 differences, with reasons
Entries
1,284
Matched
1,271
Output
differences.csv · summary.md
01Workflowfrom the chat or the editor
Month-endreconciliation
In a project with the network off: a Python step in a read-only container does the matching, and AI only writes possible reasons for what didn’t match.
Workflow · Month-end reconciliation
demo data
Month-end reconciliationSavedSaveRun
Last run · 31 October · 13 differences
02Setup and flowFinance
Setup and flow
A finance team that compares bank statements with the books by hand every month.
Setup
Project
A workspace project with the network off (no internet, only the model provider); a worker with containers for the code step.
Input and repeat
Statement and ledger export as CSV, invoices as PDF. A monthly automation set to “continue from its last run” takes the previous month’s result as input.
What a day looks like
01
At month end you drop the statement and the ledger export into the workspace.
02
A Python step matches entries by amount, date and description (in a read-only container with no network); the agent writes possible reasons for the unmatched ones.
03
In the morning differences.csv and a short summary are in the Outputs screen; the decision is the accountant’s.
What changes
Hours of line-by-line comparison become a checklist; the team spends its time on the differences that really need a look.
How it connects
This setup needs no outside accounts: in a project with the network off, files go in and a table comes out. If you want, you also connect to your accounting software’s or bank’s API with the workflow HTTP step and set the fields to fit your own setup.
We don’t impose a ready-made flow; design your own, either with AI (describe it in the chat) or yourself (in the workflow editor).
Good to know
The result is a checklist for your accountant to decide on; differences.csv opens straight in Excel.
Model calls go to the model provider you choose; for sensitive data you choose the provider accordingly.
Security settings
demo data
A screen from the console · The real console: security settings, with the Sıkı (strict) preset for sensitive work.
This example runs on the same infrastructure as all the others: Remyral on your own server, and we build your flow together in the pilot. Just describe the work you want to simplify and the tools you use in your application.
Remyral16:10
