Build with AI

Design a conversational interface that interacts with workflow canvas to build complex enterprise automations

My Role

UX Design, Prototyping, Stakeholder buy in

Timeline

6 weeks

Prompt card: Describe the automation you want to build and/or upload your process doc to start

Context

Quick Automate is a tool that automates complex enterprise processes like claims processing, appointment scheduling, compliance checking. These workflows involve conditional branches, loops, and cross-departmental handoffs that are difficult to visualize and even harder to build without errors. Traditionally, they are built by trained automation developers due to their complexity but the goal was to lower the barrier of entry to building.

The Solution: Chat-to-Workflow

A conversational interface to transform a technical chore into a simple dialogue.

Intent Recognition

Instead of mapping out 20 manual steps, a user simply describe what they want to do. The system interprets the goal and handles the background "plumbing."

Abstracting the Technicals

The chat interface hides the complexity of switching between UI and API. The user doesn't need to know how the data moves—only that it does.

Human-in-the-loop

If the AI agents encounter a vague step, they simply ask for clarification via chat, making the setup feel like a collaboration rather than a programming task.

Explore
Amazon Quick Suite
Automation projectsProjectname

Automate with ease

Tell us about your automation or get started with a sample

Describe the step-by-step process you want to automate or upload a process document to get started.

Add document

Select from existing documents or Drag and drop a file in .docx format (15 MB limit).

Get started with samples

Search samples...

Analytics

Billing

Sales

Design Tenets

1.

Enhance discoverability and navigation by enabling natural language interactions to help users explore/locate features, and navigate the tool efficiently. Chat serves as an intuitive interface that reduces cognitive load and learning curves while guiding users to the right capabilities.

2.

Save users time and effort by handling repetitive tasks that require multiple clicks to accomplish, thus make authoring workflows more efficient. Chat acts as an AI assistant that implements changes and surfaces relevant components to keep users informed of the actions taken to maintain transparency.

3.

We recognize that Gen AI is not always accurate. Hence, we design for iterative interactions, clearly communicate proposed modifications and provide intuitive options for users to accept, reject, or refine changes. The user has the ultimate control on building the automation.

4.

Meet users' expectations by incorporating familiar interaction patterns and capabilities they expect from conversational AI tools like ChatGPT. We do this while also maintaining the mental model and contextual awareness of automation development environment.

5.

Every action that user is able to take through Chat, must have an equivalent action on the canvas, thus preserving user control across both interfaces. There are no automation related actions that users can only take on the chat pane.

Architecture diagram showing the Assistant connecting user interaction and contextual output to Developer-at-your-side features (intuitive chat, text to action, hybrid authoring) and Visual playground (drag/drop, flow understanding)
Automation assistant

HELO, I added a step to raise a human review task in case the website fails compliance

  • In Step 5, Compliance Check Passed: I added a step to track the website details if it passes compliance

Step 5

Compliance Check Failed (Send to huk...

Step 6

Compliance Check Passed

Before we continue building, I suggest reviewing the details by clicking the edit step. You can also follow up by describing changes to the flow.

Review and edit instructions to generate Step 1

Step 1 is collected

4

Ask a follow up to make changes to the automation

image is sent to data Responsible AI Policy
Built with v0