inro

An AI-powered interview prep workspace that turns scattered job search materials into a focused, confidence-building practice system.

inro is an AI-powered interview prep workspace that transforms scattered job search materials into a structured, confidence-building practice system.

Timeline

January 2025 — March 2025

Role

Lead UX Designer

Tools

Figma, Claude, Cursor

Context

Most interview prep today is unstructured. Candidates collect information but struggle to turn it into a clear plan.

Most candidates walk into the room with a messy Google Doc, ten open LinkedIn tabs, and a massive amount of anxiety.

To understand this space, I interviewed candidates across different fields and asked them to walk through their real prep workflows and artifacts. I then synthesized the findings into an affinity-mapped journey, capturing user intent, role setup, AI interaction, practice, and progress tracking.

Problem

Interview preparation is fragmented and cognitively heavy.

Candidates do not lack information. They lack structure. Across interviews and artifacts, a consistent pattern emerged. People spend more time organizing and collecting than actually practicing.

Multiple tabs and resources

Notes scattered across documents

Opportunity

How might we turn unstructured inputs into a clear, actionable preparation system that builds confidence and improves performance?

Key insights

Mapping the Threat Intelligence Lifecycle

From 6 interviews, a survey with 20+ responses, and journey mapping:

Candidates over-index on organizing instead of practicing. Many spend hours refining notes they rarely revisit.

Preparation lacks prioritization. Users struggle to identify which skills or experiences matter most for a role.

Confidence is miscalibrated. Reviewing material creates a false sense of readiness without practicing articulation.

Tools are fragmented. Preparation is spread across documents, tabs, and mental models with no unifying system.

Candidates over-index on organizing instead of practicing. Many spend hours refining notes they rarely revisit.

Preparation lacks prioritization. Users struggle to identify which skills or experiences matter most for a role.

Confidence is miscalibrated. Reviewing material creates a false sense of readiness without practicing articulation.

Tools are fragmented. Preparation is spread across documents, tabs, and mental models with no unifying system.

Ideation

I created low-fidelity flows to explore how setup, the Interview Brief, practice, and history connect in one system. These wireframes validated the step-based structure and AI guidance before high-fidelity design.

User PErsona

Design principles

Chose guided structure over full automation

Fully automated prep reduced engagement and trust. Users preferred co-creating their plan.

Prioritized practice over content generation

Show key analysis steps instead of hiding them. Clearly separate inputs, internal reasoning, and outputs.

Limited surfaced insights

Every surface leads to a clear next step (from Brief to practice, from practice to suggested next session). Progress is tied to actions candidates take, not passive metrics.

Chose guided structure over full automation

Fully automated prep reduced engagement and trust. Users preferred co-creating their plan.

Prioritized practice over content generation

Show key analysis steps instead of hiding them. Clearly separate inputs, internal reasoning, and outputs.

Limited surfaced insights

Every surface leads to a clear next step (from Brief to practice, from practice to suggested next session). Progress is tied to actions candidates take, not passive metrics.

Making AI Legible

A key design challenge was avoiding a black box experience.

Each insight links back to source inputs such as resume lines and job requirements

Users can edit or refine AI-generated interpretations

System explains why something is important, not just what this builds trust and supports user agency.

The solution

inro turns raw inputs into a structured prep loop. Users upload their resume and a job description, receive an AI-generated brief with key skills and gaps, choose focus areas, and practice with tailored questions. With ongoing feedback, they continuously refine what to work on next.

Interview Brief

inro reads the resume and job description, then shows what the role is looking for, which parts of the resume are most relevant, and where there are gaps. This gives candidates a clear picture of what really matters for this interview.

Focus Selection

Candidates pick what to work on next, using suggestions for weak spots, key skills for the role, and their own confidence levels. This turns a long list of notes into a short, focused plan.

Practice Workspace

Candidates practice answering targeted questions based on the actual role. Questions are grouped into themes like behavioral or technical, and each session is set up to encourage feedback and improvement over time.

Progress Signals

Progress shows what the candidate has practiced, where their confidence is growing, and what still needs attention. Instead of vague scores, the system highlights the next best area to work on.

Key takeaways

This is an early-stage product, but I tested the concept and prototype with candidates actively interviewing.

From qualitative feedback and time-tracking exercises, I saw:

20%

Spent less time organizing notes

35%

Moved more quickly into practicing answers

3x

Reported higher confidence in what they chose to focus on

Next steps

If I had more time, I would explore a tighter integration with calendar tools to prioritize roles based on upcoming interview dates. I’d also want to refine the "Confidence Markers" into a more granular heat map of skills (e.g., "Strong on technical, weak on storytelling") to give users an even stronger signal of their growth across different applications.

Making AI Legible

A key design challenge was avoiding a black box experience.

Each insight links back to source inputs such as resume lines and job requirements

Users can edit or refine AI-generated interpretations

System explains why something is important, not just what this builds trust and supports user agency.