AgentEdge 1.0: AI Banking Platform

At FIS, I led the end-to-end design of AgentEdge, an AI-powered web platform that streamlines deposit origination for bankers and enhances customer relationships.

Project introduction

AgentEdge is a platform that helps bankers streamline client onboarding and deposit origination by leveraging AI. Instead of manually capturing client information during branch meetings, bankers can record conversations and let AI transcribe, extract, and populate data fields. AgentEdge also utilizes AI to generate meeting summaries, sentiment insights, and tailored product recommendations—empowering bankers to reduce administrative burden and strengthen client relationships.

My team

  • I led the end-to-end design process for the beta release, from discovery research through feature prioritization, wireframes, and prototypes to engineering handoff. The product is now in beta development and preparing for early-adopter testing with two bank clients.

  • In collaboration with 1 product manager & engineering team

Introducing the problem

Account opening is a critical entry point for bankers. It’s where trust is built and long-term client relationships often begin yet the process is bogged down by:

  • Manual data entry and documentation

  • Complex compliance requirements

  • Disconnected systems requiring duplicate effort

  • Pressure to meet sales targets while preserving client trust

Our team set out to design a solution that addressed these inefficiencies while keeping bankers in control. With AI as a partner rather than a replacement, AgentEdge enables bankers to focus on client relationships instead of paperwork.

Discovery & Research

Working alongside my user research partner, I began with field research and in-depth interviews to understand how bankers currently approach deposit origination.

Our objectives included:

  • Identifying decisions, challenges, and limitations in current workflows

  • Exploring expectations around AI transparency, control, and tone

  • Validating assumptions on whether AI could fit naturally into banker workflows

Insights & Opportunities

Our research surfaced five consistent pain points — each of which became an opportunity to reimagine the banker experience with AI:

1. Account openings as a relationship-building moment
Bankers see account opening as a critical entry point to form long-term client relationships, but current workflows force them to focus on administrative tasks instead of rapport.
How might we reduce administrative overhead so bankers can stay present with their customers?

2. Compliance rules creating delays and tension
Complex compliance requirements frequently disrupt the natural flow of banker–client conversations, sometimes leading to mistrust.
How might we use AI to provide subtle, real-time compliance support without undermining the banker’s authority?

3. Manual documentation frustrating both bankers and clients
Repetitive note-taking, data entry, and cross-checking slow conversations, introduce errors, and make clients feel “processed” rather than heard.
How might we streamline data capture through automation, while keeping bankers in control to review and verify accuracy?

4. Sales pressure at odds with client trust
Bankers are required to pitch products, but risk damaging relationships if recommendations feel irrelevant or forced.
How might we help bankers position product recommendations as supportive, customer-centric suggestions?

5. Fragmented systems driving inefficiency and errors
Bankers often juggle multiple disconnected platforms, causing duplicate work and inconsistent records.
How might we consolidate these tools into a single AI-assisted dashboard that integrates workflows in one place?

Design & collaboration

From here, I partnered with product and engineering to translate insights into a beta-ready solution. My contributions included:

  • Feature Prioritization: Leading sessions with stakeholders to balance user needs with engineering feasibility.

  • Design Explorations: Moving from low-fidelity wireframes to high-fidelity prototypes, testing patterns for meeting layout, AI transparency, and product recommendation flows.

  • Workshops: Facilitating ideation workshops to explore competitive features and push thinking on AI interaction design.

  • Cross-Functional Alignment: Running scoping sessions with engineering to ensure technical feasibility and seamless integration with existing banking systems.

  • QA Partnership: Collaborating closely with engineers to review builds, ensuring the fidelity of the design carried through to development.

Outcome

Beta Release

After five months of research, iteration, and cross-functional collaboration, we delivered a beta design that is currently being developed and tested with two early-adopter banks. The beta includes:

  • AI transcription & auto-populated forms to reduce manual entry

  • Banker-controlled verification flows to ensure transparency and trust

  • AI-generated summaries and notes to simplify follow-up and enhance relationship building

  • Product recommendations grounded in meeting transcripts, framed as supportive suggestions

Looking Forward

Beyond beta, I facilitated an ideation workshop and created an effort/impact analysis to shape the future roadmap. Proposed fast-follow features include:

  • Document upload with AI-powered data extraction

  • Expanded customer relationship insights

  • Further AI-enhanced coaching and compliance support