Skip to main content
Back to challenges
Challenge

Build an Audit Quality Assurance and Risk Scoring Engine with Vercel AI SDK

Construct an automated audit quality control engine for professional services firms. Using the Vercel AI SDK in TypeScript, your workflow will stream review feedback, analyze working paper logs for policy non-compliance, and flag quality risks with actionable remediation steps.

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 streaming TypeScript audit QA tool with Vercel AI SDK to flag compliance failures in accounting working papers.

Delivery guide

How work is evaluated

Evaluation

Evaluate quality risk identification accuracy on sample audit engagement files.

Datasets

Challenge datasets

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

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

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

deficiency_detection

Correctly flags missing partner sign-off defect

Binary check

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

Dimension 2

qa_scoring_accuracy

Correlation with expert reviewer QA ratings • target: 0.91 • 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

What you should walk away with

  • Implement structured streaming objects using Vercel AI SDK and Zod

  • Build automated audit paper check logic against professional accounting standards

  • Generate automated quality risk scores for enterprise audit engagements

  • Stream remediation checklists directly to partner review dashboards

Resources and assets

Reference links and supporting material

Dataset notes

Anonymized workpaper extracts and peer review audit failure records.

Challenge setup

Evaluate quality risk identification accuracy on sample audit engagement files.

Challenge input

JSON with engagement_id, workpaper_text, partner_signed (bool)

AI SDK

Vercel toolkit for streaming and structured schema outputs.

Evaluated output

JSON with quality_risk_score, deficiencies_found (array), status

Evaluation checks
  • Correctly flags missing partner sign-off defect
  • Correlation with expert reviewer QA ratings • target: 0.91 • range: 0-1
Configured evaluation evidence
  • Benchmark: Audit_Quality_QA_Bench
  • Qa Scoring Accuracy target: 0.91
  • 1 public reference case
Configured runtime
  • JavaScript execution harness
  • javascript sandbox (unavailable on Versalist)

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

Action Space
  • Vercel AI SDK · Required
  • /dev/agents · Optional
  • Metaflow · Optional
Orchestration
  • Vercel AI SDK · Required

Evaluation contract

  • deficiency_detection · Weight 1
  • qa_scoring_accuracy · Weight 1

Metrics

  • qa_scoring_accuracy · Target 0.91

Benchmarks

  • Audit_Quality_QA_Bench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON with engagement_id, workpaper_text, partner_signed (bool)

    Output: JSON with quality_risk_score, deficiencies_found (array), status

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · TypeScript Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an Audit Quality Assurance and Risk Scoring Engine with Vercel AI SDK