Skip to main content
Back to challenges
Challenge

Automating Finance Audits with Gemini & CrewAI

Develop an AI-powered auditing system for financial transactions using Gemini 2.5 Pro and CrewAI. The system should leverage Gemini's advanced reasoning capabilities to analyze financial data and identify anomalies. CrewAI will orchestrate a team of specialized agents, each responsible for a specific audit task (e.g., fraud detection, compliance checks, data reconciliation). The system should integrate with enterprise financial systems via MCP and incorporate extended thinking for complex case analysis.

Workflow AutomationHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Build a hierarchical agent system using Gemini 2.5 Pro and CrewAI for automated financial auditing.

Delivery guide

How work is evaluated

Evaluation

Evaluate the system's accuracy in detecting financial anomalies, its efficiency in processing large datasets, and its ability to handle complex cases.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Learning goals

What you should walk away with

  • Develop an extended thinking module for complex case analysis.

  • Implement Gemini 2.5 Pro's advanced reasoning for anomaly detection.

  • Evaluate system performance using real-world financial datasets.

  • Design a CrewAI agent hierarchy for financial audit tasks.

  • Integrate MCP for secure access to enterprise financial systems.

Resources and assets

Reference links and supporting material

Dataset notes

Sample data for 1 tasks

Challenge setup

Evaluate the system's accuracy in detecting financial anomalies, its efficiency in processing large datasets, and its ability to handle complex cases.

Challenge input

CSV file of financial transactions

Agent execution

No agent tool is listed for this challenge.

Evaluated output

JSON report with identified anomalies and explanations.

Evaluation checks
  • Checks if the system correctly identifies known anomalies.
Configured evaluation evidence
  • Precision target: 0.9
Configured runtime
  • Python execution harness

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

No tool records are attached.

Evaluation contract

  • Checks if the system correctly identifies known anomalies.

Metrics

  • Precision ยท Target 0.9

Workflow tasks

  1. Task 1

    Input: CSV file of financial transactions

    Output: JSON report with identified anomalies and explanations.

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Automating Finance Audits with Gemini & CrewAI