fazalehaq.comSaved
LinkedIn ↗X
A1
LAB
AIDBDRILLCTOOLDLEVELESTATUS
← LAB

Python · Professional · Free practice

Controlled Finance Reconciliation Pipeline

Run a two-source reconciliation with prematch validation, exception classes, control totals, and a human review queue — based on the PYFIN-004 Master Lesson pattern.

00GOAL

What you will produce

Run a two-source reconciliation with prematch validation, exception classes, control totals, and a human review queue — based on the PYFIN-004 Master Lesson pattern.

01PREVIEW
RAW_A = PREMATCH_EX_A + MATCHABLE_A
MATCHABLE_A = CLEAN + AMT_MISMATCH + CCY_MISMATCH + UNMATCHED_A
# Cross-source amount delta = diagnostic, not completeness

// Control identities — completeness before trust

02PREREQUISITES
  • Comfortable with pandas
  • Familiar with finance reconciliation concepts (keys, amounts, currency)
  • PYFIN-004 Master Lesson approved in Content OS (source pattern)
03PROCEDURE

Run the drill

Step sequence

  1. 01

    Validate required schema on both sources before any merge.

    Step
  2. 02

    Classify prematch exceptions: missing keys, duplicate keys, invalid amounts/currency.

    Step
  3. 03

    Match on the reconciliation key; separate clean matches, amount mismatches, currency mismatches, and unmatched rows.

    Step
  4. 04

    Prove completeness with control identities (raw → prematch exceptions + matchable; matchable → clean + mismatches + unmatched).

    Step
  5. 05

    Write an exception log and a short human-review queue for residual judgment.

    Step
04CONTROLS

Validation gates in this drill

  • Validate required schema on both sources before any merge.

    Pass

  • Classify prematch exceptions: missing keys, duplicate keys, invalid amounts/currency.

    Exception

  • Prove completeness with control identities (raw → prematch exceptions + matchable; matchable → clean + mismatches + unmatched).

    Exception

  • Write an exception log and a short human-review queue for residual judgment.

    Exception
05 EXPECTED_OUTPUT

Expected outcome

A professional drill that treats reconciliation as a controlled system — not a single merge — with completeness evidence and exception ownership.

Target state
06FILES

// Use Content OS reconciliation datasets when published with this drill