Overview

The AI Knowledge Store lets IBM Concert clients upload their own documents so the AI agent and watsonx ( IBM's enterprise-ready AI) chat can answer with their organization's knowledge. I led the UX design, and the feature is now live in the product.

IBM Concert is an IT automation platform that helps teams manage application risk, compliance, resilience, and cost. Out of the box, its AI agent only knows general information. Clients, especially in compliance, needed it to know their policies, procedures, and frameworks.

The feature covers the full lifecycle of a document:

• Upload a file (.pdf, .json, or .txt, up to 32 MB) and tag it with a subject and labels

• Test how the AI answers sample questions before and after adding the document, and pick the better response

• Manage the library: view details and history, pause a document, archive it, or delete it

I kept everyone aligned through shared flows, regular design reviews, and early prototypes. When a constraint changed, such as what the backend could generate, we adjusted the design together instead of handing work back and forth.

Designing around the document lifecycle

I started by mapping the full life of a document, from the moment a user picks a file to the day it is retired. That map broke the experience into four parts, each answering a question users had:

• How do I add a document? The upload wizard

• Will it make the AI better? Testing responses before committing

• How do I stay in control? The library and its actions

• What changed, and who changed it? Details, history, and the full document

1. A two-step upload wizard

Uploading opens a side-by-side wizard with a progress tracker: Upload and add details, then Test response results.

• File limits (32 MB; .pdf, .json, .txt) are stated up front, so errors are rare

• The document name fills in from the file name and stays editable

• Subject (such as Compliance, Resilience, Cost) and Labels (such as Procedure, Documentation, Framework) organize the library and give the AI context

• Next stays disabled until a file and name are present

2. Test Before You Commit

This is the heart of the design. The AI generates three sample questions related to the document, such as "Are my applications compliant with industry regulations, and how can I improve?" For each one, users see two answers side by side: the AI's current answer and its answer using the new document.

• Users pick the response they prefer. A clear selected state confirms the choice.

• Regenerate responses gives a fresh pair if neither looks right

• Users step through questions 1 to 3, and Finish upload unlocks only after all three

• Every choice also teaches the AI how to respond, so testing improves quality instead of just checking it

3. A library built for control

After upload, documents land in the AI knowledge store library with a success notification.

• A table shows name, subject, labels, download, and status at a glance

• Search, filter, column settings, and refresh help admins handle large libraries

• Status (Active or Disabled) lets admins pause a document so the AI stops using it, without losing it

• Archive moves documents out of the way while keeping them recoverable in the Archive tab

• Bulk actions let admins delete, archive, or download many documents at once

4. Details, history, and the full document

Selecting a document opens a side panel with its description, size, uploader, dates, an Active toggle, and a document history log of every change. View full document opens a viewer to read the content without downloading it.

The history log matters most for compliance: admins can show who added or changed a document, and when.

Final designs: the end-to-end flow

These are the final high-fidelity designs I handed off to the frontend and backend developers. The flow covers every step, from adding a document and testing its responses to managing it in the library. Each screen includes its key states, such as empty, selected, disabled, and success, so the team could build with confidence. I stayed involved after handoff to review builds and refine details until launch.

Challenges

• Making AI behavior visible. How the AI uses a document is invisible. We had to show its effect in a way non-technical admins could judge in seconds.

• Trust in a high-stakes domain. Compliance answers can affect audits. Every action needed to be clear and reversible where possible.

• Technical constraints. File types, size limits, and how responses are generated were set by the backend. The design had to work within them and explain them plainly.

• New AI patterns. Carbon had no ready-made pattern for comparing AI responses, so we had to design one that still felt consistent with the product.

Impact

The AI Knowledge Store shipped in IBM Concert and turned a generic AI assistant into one that can answer with each client's own knowledge.

• Grounded answers for compliance. Clients can bring in their own regulations, procedures, and frameworks, so AI answers reflect the rules they actually follow.

• Confidence before commitment. The response comparison lets users see a document's effect before it goes live, which reduces the risk of adding content that makes answers worse.

• Ongoing control. Pause, archive, and delete give admins a way to retire outdated knowledge without losing it.

• Audit readiness. Document history records who changed what and when.

• A reusable AI pattern. The side-by-side response comparison gives the Concert team a model for other features where users need to judge AI output.

What I learned

The best way to build trust in AI is to show its behavior, not describe it. Letting users compare answers side by side made an invisible process concrete, and it gave them a real say in how the AI responds.