Case 02 / 06Cocomply2024 — 2025AI · FinTech
Catching what field-completeness can't see.
- AI Native
- Banking Compliance
- Trust UX
- Enterprise
Stewards were burning cycles on re-submissions because the platform validated fields, not meaning. I designed an AI governance layer that surfaces regulatory semantic gaps before review.
- Role
- Design Lead, AI Design Consultant
- Timeline
- 2024 — 2025
- Team
- Founder + product + engineering
- Tools
- Figma, AI Prototyping, User Research
The deeper cut
Most of my work lives behind NDAs and embargoes. The walkthrough that matters — the why, the things I cut, the calls I'd make differently — happens in conversation, not on a page.
The design problem
How do we reduce steward re-submission cycles by surfacing regulatory semantic gaps before a submission reaches manager review?
Outcome at a glance
AIFinTech100 + NVIDIA Inception
Launched with enterprise banking clients. Recognized as a 2025 AIFinTech100 Top 100 AI in Financial Services and selected as an NVIDIA Inception Program participant.
Top 100AIFinTech100, 2025
NVIDIAInception Program
EnterpriseBanking clients live
02My contributions
Where I made the calls.
- 01Mapped all 4 certification phases to surface unstructured judgment requirements, and formalized semantic sufficiency against field completeness as the design brief.
- 02Defined input scope, output taxonomy, trigger model, and MVP constraints before any interface design — each constraint was a deliberate trust-scoping decision.
- 03Designed the governance assistant panel, the gap card information architecture, the severity hierarchy, and the re-scan affordance.
- 04Extended the CoComply design system for the agent layer and produced annotated screens with component-level rationale for engineering handoff.
