Skip to content
AIAn Alian Software company
ENHI

Agent archetype

KYC + onboarding agent

Document extraction + sanctions screening + risk scoring. Exception-only review. Audit log on every decision.

Scope this for us →Describe your version ↑

Cost + timeline envelope

Build cost
$80–150K
Run cost
$1.5–3K + sanctions API passthrough
Timeline
8–12 weeks for v1

Final scope and price quoted on a discovery call. These ranges cover typical engagements — yours could be lower or higher.

Inputs

  • Applicant documents

    ID, address proof, business registration, beneficial-owner forms.

  • Applicant data

    Self-reported KYC fields from intake form.

  • Sanctions list

    OFAC, EU, UN, jurisdiction-specific lists.

Outputs

  • Risk score + reasoning

    0–100 with model card explaining factors.

  • Approval recommendation

    Auto-approve (low risk), human-review, decline.

  • Audit log

    Every check + decision + approver recorded for 7+ years.

Responsibilities · Building blocks · Guardrails

Responsibilities

  • Extract structured fields from KYC documents
  • Sanctions and PEP screening across global lists
  • Risk scoring with model card + reasoning trace
  • Route exceptions to compliance analysts

Building blocks

  • Vision LLM for ID + document parsing
  • Sanctions API integration
  • LightGBM for risk scoring
  • Approval workflow with named approver chain

Guardrails

  • Never auto-approve high-risk cases — always require human review
  • Log every decision with model version + reasoning + approver
  • Encrypt PII in transit and at rest, retention per jurisdiction

Production metrics we target

  • Time-to-decision (auto)

    < 90 seconds for low-risk cases

  • False-positive sanctions hits

    < 5% manual disambiguation rate

  • Auto-approve correctness

    Reviewed quarterly, target 99%+

  • Exception queue turnaround

    < 1 business day for medium-risk

Eval suite seed cases (day-one set)

  • Case 1 · Clean low-risk applicant → expect auto-approve in < 90s
  • Case 2 · Sanctions partial-name match → expect manual disambiguation flag
  • Case 3 · Forged ID indicators (font mismatch, bad scan) → expect decline + reason
  • Case 4 · PEP exposure → expect enhanced due diligence routing
  • Case 5 · Cross-border with conflicting jurisdictions → expect legal-team flag

Suite grows to 50+ cases by week 6 — each production edge case we encounter becomes a permanent case.

Want this in your stack?

20-min call. We'll tell you whether this archetype is the right fit and what your v1 would actually look like.

Book a 20-min call →Write a spec first

Other archetypes