Treasury Router advisor dashboard showing idle-capital opportunities and a client case workflow
All work

AI financial workflow

Treasury Router

A governed advisor workspace that surfaces modeled idle-capital opportunities, stages client cases, filters approved routes, prepares client materials, and keeps execution behind human approval.

StatusWorking advisor prototype
Primary use caseAI financial advisor workflow
My roleProduct strategy, workflow design, UX/UI, recommendation architecture, report design, prototyping, and x402 testnet implementation.
Treasury Router advisor dashboard showing idle-capital opportunities and a client case workflow

The customer problem

VIP advisors need a faster way to identify idle balances, compare controlled product routes, explain tradeoffs, document approval, and hand a clean packet to operations without giving an AI agent authority over client funds.

Turns an AI financial workflow idea into a working advisor surface that stakeholders can inspect, test, and discuss without giving the AI authority over funds.

What I built

A product workflow that makes the next decision clearer.

  • Mapped the advisor journey from opportunity detection through client approval and operations review.
  • Kept suitability, allocation, and eligibility logic deterministic while using Gate AI to explain and prepare materials.
  • Separated client-facing recommendations from internal operations packets and execution boundaries.
  • Added an x402 testnet proof-of-flow for paid, permissioned access to a structured recommendation package.

Working flow

From input to controlled action

  1. Detect a modeled idle-capital opportunity
  2. Apply client constraints and route-governance rules
  3. Compare recommendation paths and rationale
  4. Prepare advisor and client materials
  5. Record human approval and create an operations handoff

Inspectable evidence

What works in the current build

  • Working multi-step advisor product foundation
  • Deterministic route filtering and recommendation logic
  • Client packet and internal operations report generation
  • x402 testnet merchant access and human-review work-order flow
AI product strategyNext.jsTypeScriptx402Financial UXHuman approval

Operating boundary

What this case study does and does not claim

Built while working with Gate AI. The public material shows a working prototype using deterministic development data. It does not use live client funds, a live Gate execution API, or autonomous execution.

This employer-context case study demonstrates product and solution-architecture capability. It is not offered for acquisition or independent licensing.

Relevant problem?

Build the version your customers can actually test.

Share the current workflow, user, and business constraint. I will tell you where I would start.

Discuss a related build