Home

Work

PRODUCT MANAGER ACCELERATOR

Turning Interview Anxiety into Confidence: Redesigning the AI Interviewer for PMs

This is an ongoing project, the case study will be updated at the end of February.

Overview

Practicely is an AI-powered interview preparation platform designed to help Product Managers get ready for interviews, by practicing Behavioral and Product Sense questions and receiving feedback.

Role

Product Designer

Responsibilities

End-to-End UX & UI Design Process and Deployment

Collaborators

Product and Engineering

Timeline

Sep 24 - Ongoing

PROBLEM TO SOLVE

The existing interface is lacking structure and visual hierarchy, limiting its value for users and the business.

The need for a redesign

A quick heuristic evaluation of the existing experience revealed unclear visual hierarchy, spacing is inconsistent, and the layout feels unbalanced, resulting in an interface that isn’t intuitive to navigate.

BUSINESS NEED

An interface that’s easy to scale and to develop.

USER NEED

A simple, intuitive, and easy to navigate interface.

Current product: https://interviewer.pmaccelerator.io/login

RESEARCH

Competitive Analysis

Key findings

Before starting the design process, I met with Product and Engineering and asked about limitations and tried understanding what was feasible. From our conversation I learned that the biggest limitations were time and resources. The deadline being November 17th and with one front-end engineer on the team, I crafted a plan.

Exponent
Scoring hallucinations
Unknown question bank size
No transparency in evaluation criteria
Unvalidated model accuracy
Final Round
Gimmicky company voices
Hard to navigate interface
Ethical issues
RocketBlocks
No interactive AI interviewer
No real-time conversation or follow-up questions
Weak community with minimal peer interaction

IDEATION

Developing Concepts

Game plan

EASY FIXES FIRST
I prioritized quick, high-impact fixes first to unblock engineering early, enabling parallel work-streams while I focused on more complex flows.
CONSTANT COMMUNICATION
I established a culture of proactive communication across design and engineering, implementing weekly stand-ups and async workflows to unblock work faster.
REALISTIC EXPECTATIONS
I defined realistic deliverables and timelines, incorporating buffer time to account for blockers and unexpected risks.

Spacing and visual hierarchy
I tackled spacing and visual hierarchy issues in the Settings first. I created a balanced layout with defined sections that are both easy to navigate and create less cognitive load.
BEFORE
AFTER
Creating the “Interviews” hub
When I tested the product, I noticed that I had to jump between a page for interview transcripts and another for feedback summaries. This created unnecessary friction and made it harder to quickly compare learnings across sessions.

To streamline the workflow, I consolidated both into a single Interviews page. Here, users can see their full interview history at a glance, and for each interview, access both the feedback and transcript side-by-side in one place.

This reduces navigation overhead, supports faster pattern recognition, and makes the learning experience feel continuous rather than segmented. By unifying these surfaces, the product now better supports PMs in doing what they came to do: learn efficiently and make informed decisions.
BEFORE
AFTER
Color contrast and accessibility
The original interface relied heavily on bright gradient panels and saturated accent colors, which created visual noise and made it difficult for users to focus on the content itself. Important information often competed for attention making it hard to focus on what matters.

To address this, I refined the color system with accessibility in mind. I introduced more neutral background tones, softened the accent colors, and applied them more sparingly to highlight only key actions or insights. Text contrast was increased to meet WCAG guidelines, ensuring the content is comfortable to read across different screens and lighting environments.

These adjustments created a cleaner, calmer visual environment. The experience now feels more balanced and easier to navigate, allowing users to focus on learning from their interviews rather than working to interpret the interface.
BEFORE
AFTER
Next Steps
- Testing with users
- Shipping the new version on February 17th
This is an ongoing project, the case study will be updated at the end of February.