QMentis.ai: Making GenAI-powered QA testing visible, fast, and trustworthy.
QMentis.ai is a GenAI-powered QA testing tool built to automate test creation and reduce timelines. By the time I joined, the tool was live, but no one could tell if it was working. I led UX for the next phase: a central operations dashboard to give decision-makers the clarity and control they lacked.
UX & UI Design
2 months
6 crossfunctional teammates
QualiZeal is a US-based enterprise quality engineering company whose GenAI platform, QMentisAI, automates software testing to cut QA timelines by up to 60%.
QMentis: AI for better testing
QMentisAI was live. The next phase was proving it worked. I joined to lead the design of a central operations dashboard that made AI performance visible for the first time.
IMPACT
What changed because this shipped?
After release, Mentexa had a single source of truth for operational health, clear role-based visibility without overwhelming anyone, and faster alignment between leadership, ops, and execution teams.
DISCOVERY
Starting mid-build with no clear brief
There were no clear requirements for what the dashboard needed to do. So I created them.
-
Interviewed the support team, who were handling confused calls from users and admins to map pain points and understand the system
-
Synthesized needs across PMs, engineers, and client stakeholders
-
Did a quick competitive scan of platforms like Salesforce and Asana to benchmark visibility features
RESEARCH
What I heard when I started listening
CHALLENGE
As QA work scaled through automation, visibility didn’t keep pace.
Leaders needed a way to see, trust, and act on real-time work, without slowing teams down.
RESEARCH
Why Qmentis needed a system, not just screens
This wasn't just a design task; it was about creating a measurable foundation for an advanced AI product.
LEADERSHIP NEEDS
SIGNALS REQUIRED
VISIBILITY GAP
SYSTEM I DESIGNED
Identify teams that need support
Reward top performers
Reassign work across teams
Track token usage and forecast needs
Measure ROI of AI testing
Understand adoption across orgs
Trust AI output for decisions
Team performance and workload
User activity and output
Ownership and team mapping
Token consumption
Usage vs Efficiency
Active vs Inactive users
Explanable insights
No org wide visibility
Performance buried in logs
No structured team layer
Invisible across org
No proof of value
Fragmented data
AI felt like a blackbox
Org - Team - Users
Performance Metrics
Team assignment controls
Token visibility at all levels
Efficiency and Impact in dashboard
Centralised usage tracking
AI insights layer
Turning complexity into something leadership could read
The challenge wasn't just adding information, it was editing ruthlessly.
I designed a central operations dashboard that surfaced decision critical information using hierarchy and balanced depth with clarity.
FINAL DESIGNS
Central Operations Dashboard
Efficiency & Impact
Centralized Usage Tracking
AI Insights
Token Visibility
Performance
-
The goal of this dashboard was to provide all key information on a single screen, prioritized and organized in a way that made sense to leaders.
-
This key dashboard solves the major challenges of providing visibility into the org and building trust in the new Gen AI testing model.
-
The data is nested in hover modes to prevent information overload. This is the main tool for leaders now to take informed actions.
FINAL DESIGNS
User Role Management
Org - Team - Users
Team Assignment Controls
AI Insights
-
Allows leaders to oversee information for each user and assign or edit their roles and teams.
-
Leaders can view data of active and inactive users according to time, roles, and services.
.png)
FINAL DESIGNS
Projects and Teams Management
Org - Team - Users
Team Assignment Controls
AI Insights
-
Allows leaders to oversee information for each team, their performance, and the users assigned.
-
Leaders can view data of active and inactive teams according to time, roles, and services.
-
Teams can be edited, and users can be assigned to different teams if needed.
PROCESS
Designing while shipping (Iteration in real-time)
This project didn’t happen in a vacuum. I was designing while engineering was actively building, requirements were evolving, and feedback was coming in mid-sprint. That meant tighter loops, faster decisions, and constant tradeoffs. Not everything could be perfect — but everything had to be intentional.
RESULTS
What changed after release
MVP had immediate impact
Below is a testimonial from one of the current users of QMentis.ai
Teams had a shared view of operations,
less back-and-forth and more confidence in making decisions.
This wasn’t about aesthetics. It changed how decisions were made and how fast teams could respond.
LEARNINGS
What this project taught me
-
Adapt quickly and talk about the challenges to stakeholders and the team before designing.
-
Calling the shots in ambiguity and working with the PM and engineers to work faster and find solutions.
-
This was a small part of a very complex software. It taught me how to design with scalability in mind and move fast by making trade-offs and prioritizing pain points. Sometimes, you have to trade off pixel perfection for problem-solving.
.jpeg)
