Skip to main content
Back to challenges
Challenge

Build an Authenticated Corporate Onboarding Agent with Google ADK and Stytch

Onboarding Singapore corporate banking entities requires multi-step authentication and business register validation. Construct an agent using Google ADK and Stytch authentication services to parse corporate filings, verify authorized signers, and score onboarding risk.

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 multimodal identity validation pipeline using Google ADK and Stytch to automate corporate entity verification and risk scoring for Singapore business onboarding.

Delivery guide

How work is evaluated

Evaluation

Evaluates document processing accuracy and Stytch identity verification tool calls for corporate account opening.

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

verifications_completed

Ensures both UEN and Stytch verification steps were performed

Binary check

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

Dimension 2

risk_scoring_accuracy

Absolute error against target risk benchmark • target: 0.95 • 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 multi-agent workflows using Google ADK and Gemini multimodal models

  • Integrate Stytch identity verification APIs into agent decision tools

  • Extract structured entity records from unstructured ACRA corporate PDFs

  • Implement automated KYC risk profiling compliant with MAS guidelines

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic ACRA business profiles and Stytch auth session payloads.

Challenge setup

Evaluates document processing accuracy and Stytch identity verification tool calls for corporate account opening.

Challenge input

JSON object with company_uen, director_email, and document_url

Google ADK

Provides Gemini agent execution harness

Stytch

Performs corporate identity verification

Evaluated output

JSON object with uen_valid, stytch_user_verified, overall_risk_score, and onboarding_decision

Evaluation checks
  • Ensures both UEN and Stytch verification steps were performed
  • Absolute error against target risk benchmark • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: KYC-Agent-Bench
  • Risk Scoring Accuracy target: 0.95
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python 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
  • Google ADK · Required
  • Stytch · Optional
  • Google · Optional

Evaluation contract

  • verifications_completed · Weight 1
  • risk_scoring_accuracy · Weight 1

Metrics

  • risk_scoring_accuracy · Target 0.95

Benchmarks

  • KYC-Agent-Bench
  • security

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with company_uen, director_email, and document_url

    Output: JSON object with uen_valid, stytch_user_verified, overall_risk_score, and onboarding_decision

    Execution harness: Python

Evaluation tools

  • Google ADK · Agent Framework
  • Stytch · Authentication Tool

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an Authenticated Corporate Onboarding Agent with Google ADK and Stytch