BrightHive sought to democratize their data platform by integrating AI agents that make complex analytics accessible to non-expert users. We defined comprehensive user flows and workflows for different personas, designed an innovative AI agent interaction model, and created the UX/UI that brings it all together.
The result is a platform where specialized AI agents collaborate through conversational interfaces, handling complex data tasks while maintaining transparency. This approach transforms advanced data analysis from an expert-only capability into something accessible to anyone through natural language interactions.
Product Design
Userflows & workflow definition
Interaction model
UX and UI design
The centerpiece of our solution was a hierarchical AI system where specialized agents work together under "The Queen" coordinator. This architecture handles complex tasks behind a simple interface, making advanced data operations accessible to non-technical users.
Collaborative sessions with the BrightHive team to map essential user flows. Each colored sticky represents different user actions or system responses, creating a visual journey map for non-technical users navigating the platform with AI assistance.
Screen-by-screen mapping of critical processes, showing exactly how users progress from question to insight with BrightBot's help. This detailed approach ensured we maintained technical requirements while simplifying the experience for non-data experts.
Our iterative design process explored multiple interface solutions for the AI chat experience. Through these variations, we tested different patterns—from drawers to panels to step navigators—to find the right balance of simplicity and functionality that would serve both technical and non-technical users.
Early digital mockups exploring different approaches to the AI chat experience. These designs tested various concepts for displaying collaborative AI agents, integrating data visualizations within conversations, and maintaining context as users move through complex analysis workflows.
The design evolved through multiple iterations, from drawer-based navigation to panels and step navigators, each refining the experience for different user types and workflows.
We redesigned BrightBot's interface to balance simplicity with powerful functionality. The step navigator guides users through processes while the collapsible panel provides detailed views without clutter. Smart loaders show AI agents working in real-time, and integrated file management enables seamless data upload and result downloads.
BrightBot creates detailed plans using specialized agents when handling complex requests. This dual-view interface reveals the AI's process, building trust by showing users both the planned approach (left) and real-time execution (right).
This transparency serves two key purposes: it lets users understand and modify the process at any point, while educating them about data operations that would typically require technical expertise.
One of BrightBot's most powerful features is letting you ask questions about your data in plain English. Simply type "Show me last month's sales by region" or "Which customers bought the most?" and get instant answers.
The system translates natural language into technical queries behind the scenes, enabling anyone to explore data through conversation while maintaining accuracy for technical teams. This approach removes barriers between business users and their data insights.
Our dual-mode interface adapts to each user's technical comfort level. The "Preview Data" view (left) presents insights in business-friendly formats with explanatory text, while the "See Code" toggle (right) reveals the technical implementation for those who want to verify or learn from the process.
This approach allows the platform to serve both technical and non-technical users without compromise.
The redesigned BrightBot experience has fundamentally transformed how users interact with the platform. By making advanced data analysis accessible through natural conversation, BrightHive has evolved from a data tool into a powerful data partner. The AI agent interface has increased platform engagement and generated new interest from potential users, while unlocking powerful capabilities that were previously limited to technical experts. This shift represents a new era where anyone can gain meaningful insights from complex data through simple dialogue.
