Crebot — Shivani Mistry

An AI conversational tool that transforms natural language prompts into fully editable, customizable dashboards — designed for enterprise fintech teams who needed speed without losing control.

Timeline

Mar – Dec 2024

My Role

Product Designer

Methods

User Research · Prototyping · Usability Testing

Outcome

70% faster dashboard creation

Leveraged Skills

Figma · UX Research · AI Interaction Design · Prototyping

Opportunity

Manual dashboard creation was costing teams hours.

Enterprise dashboard creation is often manual, fragmented, and time-intensive — requiring users to configure data, structure layouts, and define metrics with technical precision. Non-technical stakeholders were dependent on engineers for even basic reporting changes.

Crebot was built to transform this into an intuitive conversational experience while maintaining the accuracy, control, and trust that enterprise teams demand.

How it works

From prompt to dashboard in seconds.

Crebot interprets natural language, maps intent to structured data components, and generates an editable dashboard in real time. Users can then drag, resize, and refine every element — keeping full ownership of the final output.

Concept Exploration

Mapping the problem space.

Early exploration mapped the core pillars of the product: identity & tone, customization, data input, widgets, and command & conversation — grounding the design in user needs before moving to screens.

crebot-concept-exploration.png

Drop your concept sketch here

Design Principles

Three principles that shaped every decision.

Automation accelerates, user stays in control

AI should reduce manual dashboard setup while keeping users in control of structure and data. Every generated output is a starting point, not a final answer.

Editability preserves ownership

Generated dashboards remain fully customizable to maintain user confidence. If a user can't change something, they won't trust it.

Conversation that guides, not dictates

The interface supports flexible prompting while offering intelligent suggestions to reduce ambiguity and help users get to the right output faster.

Outcome

Shipped. Measured. Validated.

70%

Reduction in dashboard creation time after launch.
Deployed to enterprise fintech teams across the platform in late 2024.

Some details anonymized due to NDA.

Key Screens

Core experiences, designed end to end.

crebot-keyscreen-01.png

Getting Started

Users begin by describing what they want to create. Crebot interprets natural language prompts to generate a structured dashboard instantly.

crebot-keyscreen-02.png

Prompt-driven Widget Creation

Natural language converts into relevant widgets that users can drag and arrange to build a fully customizable dashboard layout.

crebot-keyscreen-03.png

Drag & Drop Edit Mode

Users drag widgets to the edit canvas and customize each element's size, color, and graph style to match their needs.

crebot-keyscreen-04.png

Real-time Data & API

Live data integrations keep dashboards current. Users connect APIs and see their metrics update in real time without manual refresh.

crebot-keyscreen-05.png

Multi-dashboard & Collaboration

Users can simultaneously work across multiple dashboards and collaborate with teammates in real time, sharing views across the organization.

crebot-keyscreen-06.png

Dark / Light Mode