Accern. Rhea Design Case Study
the project
Challenge:
Kumesh Aroomoogan, a former research analyst for Wall Street firms and co-founder of Accern set on a mission to solve the problem of unstructured financial data – a facet often overlooked and inefficiently analyzed in historical data decisions.
Lazarev. engaged in the design of AI-powered solutions for Accern back in 2018.
Fast forward to 2023, Kumesh approached our team to develop a sophisticated tool for financial researchers, leveraging the capabilities of a pre-trained AI model that extended beyond conventional chatbot functionality.
Approach:
We followed the product-centered approach when prototyping Rhea, conducting deep UX research, and learning the daily workflows and problems of financial analysts.
The solution that emerged combines a prompt-driven AI interface with dynamic widgets, shortcuts, and intelligent suggestions, offering a seamless research-to-report workflow with a handy split screen mode.
The Project’s
Discovery Phase
Rhea’s functionality
At the heart of Rhea's capabilities is a prompt-based functionality enhanced by shortcuts, autocomplete, and widgets. To add transparency, we strategically placed references alongside the AI's responses, guiding users to the sources of the pulled data.
The platform is designed to seamlessly handle automated emailing, schedule calls, issue alerts, and perform various other actions integral to the day-to-day workflow of a financial researcher.
Creating an adaptive communication system
We engineered an adaptive natural language communication system. When users struggle to express their questions, Rhea steps to refine and amplify communication.
It leverages techniques such as clarification, suggestions, hints, and prompt components, breaking down the barriers of articulation to effectively guide and assist users.
Power of Digital Product Design for Series A Funding
Introducing the split screen creator mode
After closely observing how financial researchers go about their tasks, we devised the split screen creator mode in Rhea's interface to enhance usability. This innovative feature allows users to seamlessly populate a research report while continuing their research, eliminating the need to switch between tabs or manage multiple windows.
Designing the multi-purpose input field
Rhea goes beyond natural language understanding, covering user comprehension, personalization, file management, and task assistance. This comprehensive suite is supported by robust data processing and synthesis on the backend.
At Lazarev, we committed to actualizing this functionality by transforming the chat field into an advanced search bar and a command line.
Designing a hybrid GUI & prompt-based interface
Given the complexity of Rhea's functionality, we purposefully crafted a hybrid GUI/prompt-based interface.
The interface is dynamic, with buttons and widgets adapting to the context of the conversation. For instance, Rhea can showcase references, charts, footnotes, graphical controls, and quick action buttons.
Linking datasets & managing files
Within the settings, financial researchers can choose from pre-configured data sets known as Lenses or link their own datasets through Accern's NLP Platform. We provided them with the option to upload and manage files and folders, offering essential context for Rhea's data analysis.
Moreover, we crafted custom email notifications to send alerts and reports on designated topics, keywords, industries, or any other information that researchers consider valuable.