Skip to main content
Loading navigation
Back to challenges

Private Equity AI Risk Audit System with LlamaIndex RAG

Private equity portfolios face significant unmonitored regulatory and operational risks in their deployed AI payment architectures. Build a compliance auditing agent using LlamaIndex to query complex governance documents, evaluate system logs against regulatory frameworks, and achieve 90% precision on compliance gap detection.

CybersecurityHosted by Vera
Sign in free to start

What you are building

Develop an advanced RAG and auditing agent using LlamaIndex to identify governance and compliance blind spots across private equity tech portfolios.

How work is evaluated

Evaluation

Evaluates precision and grounding accuracy of compliance audit summaries generated across PE portfolio docs.

Datasets

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

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

citation_grounding_check

Verifies that all reported regulatory gaps contain valid document citations.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

gap_detection_precision

Precision score of true compliance vulnerabilities detected. • target: 0.9 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

  • Construct multi-document index structures using LlamaIndex for complex regulatory frameworks

  • Build auto-retrieval query engines with metadata filtering for portfolio governance checks

  • Detect compliance gaps between deployed AI payment systems and EU AI Act / SEC guidelines

  • Generate prioritized risk mitigation roadmaps with verifiable source citations

Resources

Dataset notes

Synthetic enterprise document repository containing 100 architecture specs, AI policy documents, and system audit logs from 10 fintech portfolio companies.

Challenge setup

Evaluates precision and grounding accuracy of compliance audit summaries generated across PE portfolio docs.

Challenge input

JSON input specifying portfolio company ID, document index key, and target regulatory framework

LlamaIndex

Framework for building data-driven LLM applications and agentic RAG.

Evaluated output

JSON audit result listing detected risks, citations, severity, and compliance score

Evaluation checks
  • Verifies that all reported regulatory gaps contain valid document citations.
  • Precision score of true compliance vulnerabilities detected. • target: 0.9 • range: 0-1
Configured evaluation evidence
  • Benchmark: LegalRAG-Audit-v1
  • Gap Detection Precision target: 0.9
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • Llama Index · Required
  • LlamaIndex · Optional
  • Llama 3.2 · Optional

Evaluation contract

  • citation_grounding_check · Weight 1
  • gap_detection_precision · Weight 1

Metrics

  • gap_detection_precision · Target 0.9

Benchmarks

  • LegalRAG-Audit-v1
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON input specifying portfolio company ID, document index key, and target regulatory framework

    Output: JSON audit result listing detected risks, citations, severity, and compliance score

    Execution harness: Python

Evaluation tools

  • LlamaIndex · Data & RAG Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Private Equity AI Risk Audit System with LlamaIndex RAG