HOME
BACK TO TOP

eigenmind.ai web app

AI-POWERED COMPANY KNOWLEDGE BASE CHAT

client
wattx venture
role
Product designer
completion
7 months, 2024
User research
UX UI Design
Design systems
Design vision
👁 TL;DR

Project setup

team
Product manager, Business analysts, Backend developer, Frontend developer, 🙋Designer
tools
Miro, Jira, Figma, Adobe Photoshop, Adobe Illustrator, Webflow

The challenge

We set out to design and prototype an MVP for a secure, locally hosted generative AI chat app trained on company data. The goal was to help teams quickly access siloed company information, get summaries, and collaborate more efficiently while meeting the strict security and data control standards expected by German SMEs.

To shape the concept, we ran user and buyer interviews, validated key assumptions, and launched a landing page to attract early interest and gather feedback. These insights guided the development of a focused MVP to test core features and demonstrate real value to potential users.

My responsibilities

📌 Support the research team in explorative research and insight review to uncover opportunities
📌 MVP vision and feature prioritization in collaboration with the Product Owner functionalities and write specs
📌 Building and maintaining a scalable atomic design system
📌 Documentation & developer hand-off
📌 Logo, branding & CI

The outcome

We successfully developed a working MVP of the AI-powered chat web app that met the core goals of our project. Additionally, the landing page campaign generated initial leads and user interest, providing a valuable pipeline for further development and testing.

Unfortunately, the project has been discontinued due to changed priorities.
01
Research

Explorative user research

Understanding the problems and painpoints through qualitative interviews

After we have established the initial concept to validated, we have focused on collecting insights from potential users and ran exploratory interviews to identify pain points and their perceived severity based on a role / company.  

Our business development team has reached out to potential users on Linkedin, trade fairs and also used personal connections to get first-hand insights.

🚩 PROBLEM
German SMEs struggle with knowledge management, knowledge transfer and access to relevant company information as the data is stored disorganized way or is not easy to find when one does not know where to look for it.
HYPOTHESES
➔ Users see value in having a single system to access all company knowledge spread across different tools, under the premise there is access management
➔ Employees are having hard time finding the information they are looking for and spend a lot of time doing that.
➔ There is need for automation of repetitive tasks based on historical data
➔ Industries like machine engineering, law, software show most interest
02
IDEATION & SCOPING MVP

Organizing insights and scoping out the MVP

Use cases, user flows, feature priorization and technical feasibility

After aggregating the interview insights, identifying use cases and In a team workshop we have sketched out a journey map and outlined the feautures every step of the process may contain. This helped us create a visual representation of

Core feautres for regular user

Based on the pain points identified during talking to the potential users we have organized a team brainstorming session to throw together all use cases and possible features regardless of their technical feasibility, cost, effort and priority just to have a library of functionalities to prioritize for our MVP.

Feature scope

Once we had a pool of functionalities, we decided to narrow down the scope and estimate which features are must-haves, which are nice-to haves and which need more validation to even begin the estimation.

For the "regular" user we have limited the scope to core chat functions, history & favorites, collaborative chat and managing local user-specific data sources.

In the next step I have organized the core features in a flow chart format to visualize the logic and move onto the next step of creating first lo-fi wireframes.

03
Prototyping & VALIDATion
04
VISUAL LANGUAGE

Typography, color scheme, iconography & layout

I have picked 2 font styles for header and body text and defined the core color palette with 2 base colors used for text and backgrounds taking up around 80% of the real estate and 2 accent colors for the 10-20% left.

I have organizes core, accent, and semantic color palettes with hex values and accessibility contrast ratios aligned with WCAG standards (AA/AAA).

The structured approach ensures consistency across the interface while supporting inclusive, accessible design for a wide range of users.

With some extra time on my hands I sketched out several modular illustrations for the landing page to add the product some personality.

05
refining the design

High-fidelity prototypes

User story based hand-off ready prototypes

After presenting the test results in the round, we carried out a workshop with the Product ownder, data scientist, developer & UI designer to ensure the implementation of the feedback is feasible and is within our timeline and budget.

Prototypes
06
DESIGN SYSTEM

Atomic design system for dev-friendly hand-off and accessible maintenance

I joined the project after initial scoping and early development had been completed by another designer, who had drafted the first version of the Savings feature and scheduled user interviews. After onboarding, I stepped into the interview phase to deepen my understanding of the target audience, support analysis, and familiarize myself with the product ecosystem.

07
CORE FEATURES

Related projects