Build an Aadhaar-Compliant eKYC Fraud Mitigation Pipeline with Pydantic AI
Digital public infrastructure identity workflows require zero-trust security and strict data privacy compliance. In this challenge, you will develop a type-safe Aadhaar eKYC verification and spoofing detection pipeline using Pydantic AI. The agent processes anonymized eKYC XML/JSON payloads, verifies digital signatures, detects biometric spoofing indicators (liveness score, facial match index), and outputs fully validated compliance certificates while guaranteeing PII masking.
What you are building
The core problem, expected build, and operating context for this challenge.
Build an eKYC fraud verification agent enforcing Pydantic AI models for zero-trust identity verification.
How work is evaluated
Evaluates PII masking correctness and eKYC verification approval outcomes.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
pii_protection_test
Verifies raw UID is never present in final verification output.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
verification_accuracy
Accuracy in correctly passing valid KYC attempts and rejecting spoofed attempts. • target: 1 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Construct Pydantic AI validation agents for eKYC payload verification.
Implement automated PII masking and vaulting procedures for Aadhaar numbers.
Evaluate facial liveness scores to prevent identity spoofing attacks.
Generate compliance verification tokens with tamper-proof signatures.
Reference links and supporting material
Synthetic eKYC payload log stream with simulated biometric match and signature verification status.
Challenge setup
Evaluates PII masking correctness and eKYC verification approval outcomes.
Challenge input
JSON with masked_uid, liveness_score, face_match_score, signature_valid.
Pydantic AI
Required framework for validated KYC pipelines.
Evaluated output
JSON with kyc_status, masked_uid_output, risk_score.
- Verifies raw UID is never present in final verification output.
- Accuracy in correctly passing valid KYC attempts and rejecting spoofed attempts. • target: 1 • range: 0-1
- Benchmark: eKYC Fraud Prevention Benchmark
- Verification Accuracy target: 1
- 1 public reference case
- 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
- Pydantic AI · Required
- Exa · Optional
- Metaflow · Optional
- Pydantic AI · Required
Evaluation contract
- pii_protection_test · Weight 1
- verification_accuracy · Weight 1
Metrics
- verification_accuracy · Target 1
Benchmarks
- eKYC Fraud Prevention Benchmark
- security
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON with masked_uid, liveness_score, face_match_score, signature_valid.
Output: JSON with kyc_status, masked_uid_output, risk_score.
Execution harness: Python
Evaluation tools
- Pydantic AI · Type-Safe Agent Framework
Recipe freeze marker
No recipe freeze time is recorded.