MONSTERCAREERS

AI Job Seeker Prototype

An AI-driven job search experience that turns user input into a personalized path to the right role.

PROJECT OVERVIEW

A hi-fidelity prototype exploring how AI could improve the job seeker experience for a newly merged Monster + CareerBuilder platform.

With the merger, there was an opportunity to rethink the job search experience from the ground up—not just improve it incrementally.

The goal was to explore how AI could:

Help users quickly establish or refine their professional direction
Generate a strong, usable resume
Deliver relevant job recommendations with minimal effort

Rather than relying on assumptions, we used rapid prototyping and user testing to define what this experience should be.

ROLE

UX/UI Designer

Designed end-to-end experience and flows
Created all wireframes and prototypes (Axure → Figma)
Moderated user testing sessions
Collaborated with Product and Engineering on AI behavior and interaction patterns

PROJECT LENGTH

1 Month

TOOLS

Axure

UXPressia

Office Suite

UserTesting.com

METHODS USED

User Interviews

Iteration

Rapid Prototyping

User Testing

Hi-Fi Prototyping

IMPACT

Validated a new AI-driven product direction for job seekers
Influenced roadmap as a net-new product concept, not a replacement
Demonstrated the team’s ability to move from ambiguity to clarity in weeks
Secured leadership buy-in prior to company acquisition

The project was ultimately shelved post-acquisition, as the acquiring company had a similar product in market.

EXPLORATION & DISCOVERY

We approached this project with a simple assumption:

AI could help users figure out what they want to do.

We were wrong.

Through moderated sessions (10 users per round via UserTesting.com), we learned:

Users don’t come to job platforms to explore—they come with intent
Open-ended, aspirational questions slowed users down
Speed and clarity mattered more than flexibility

This fundamentally shifted the product from “career discovery” to “career refinement.”

Another key insight:

Users didn’t want a single AI interaction model. They preferred a hybrid approach depending on context:

Structured inputs when unsure
Conversational flows when providing detailed information

UX/UI DESIGN PROCESS

Approach

We designed the experience as a guided flow:

Onboarding
AI asks structured questions to understand the user’s goals and background
Profile + Resume Creation
A conversational interface gathers experience, education, and skills
AI generates or improves a resume based on best practices
“Lightning Round” Job Rating
Users quickly review and rate 10 jobs (like / don’t like)
This provides implicit preference data
Refinement
AI surfaces observations:
- “You preferred remote roles…”
- “You skipped software positions…”
  Users confirm or adjust
Personalized Job Recommendations
A final job list tailored to both explicit and implicit inputs

AI INTERACTION DESIGN

We tested multiple interaction patterns and landed on a hybrid model:

Structured inputs early to reduce friction and decision fatigue
Conversational UI for resume building, where users needed flexibility

The key decision was not choosing one AI pattern—but matching the interaction model to the user’s mindset at each step.

Rather than exposing full control over AI outputs, we:

Tuned tone and length based on user feedback
Prioritized clarity and usefulness over customization

Trust was built through:

Iterative internal testing with Engineering
Tight feedback loops across design and development
Keeping outputs predictable and grounded

ITERATION

We completed three rounds of prototype testing, each built as a clickable experience:

Round 1 → 2
Reduced scope after learning users didn’t want career discovery
Shifted focus to speed and refinement
Round 2 → 3
Moved from mobile-first to desktop-first
Users strongly preferred desktop for resume creation and profile setup

Each round helped clarify what mattered most: getting users to relevant jobs as quickly as possible.

CONTRAINTS

1-month timeline from concept to validated prototype
Unclear stakeholder landscape due to merger
Undefined product direction
New and evolving LLM capabilities within the organization
Ongoing internal debate around product strategy

To move forward, I pushed for a user-led approach—using rapid testing to resolve ambiguity instead of internal debate.

FINAL DESIGN

The final prototype combined structured guidance with conversational flexibility:

A streamlined onboarding flow focused on speed
AI-assisted resume creation with contextual prompts
A fast, intuitive job rating system (“Lightning Round”)
Personalized recommendations based on real user input

The experience emphasized clarity, speed, and usefulness—helping users move quickly from intent to opportunity.

FINAL PROTOTYPE

The final prototype was created based on multiple rounds of user feedback and went through many iterations. I then used it to shop our new product concept around to company stakeholders in engineering, sales, marketing and other product groups.

I focused on new features that appealed to the small business userbase, and shared feedback from testing to back up my choices.

Break it up: A multi-page job description flow that broke the process up into logical, bite-sized bits
The Hiring Boost: many of the items in the job posting flow are optional, and likely skipped by users. We pulled user data and inserted it as a point of interest to let users know that though optional, that piece of info will yield more and better applicants.
Strength Meter: This measures how effective the job listing will be, giving real-time feedback as the user adds more data.
Candidates in Your Area: Similar to the Strength Meter, run a real-time count of candidates that Monster knows would be a good match for the job listing in its current state. The better the listing, the more candidates that match.
Location Feedback: To help employers understand why they might not be getting as many candidates as they hoped, we visualized where Monster knows candidates to be, based on the location that the employer listed. We would point out potential issues like long commutes, or a known shortage of talent in their area.
Recommended Changes: I envisioned using AI to evaluate job descriptions in real time, and rate them as “Critical” or “Important” fixes. The UI would operate like Grammarly or similar features, underlining the problematic sections, and offering suggestions upon rollover.
Save for Later: Seems obvious, but not all sites include this. Knowing that small business owners don’t have much time, we included the ability to save job listings as drafts to finish later.
Game-ify Profile completion: Users are more likely to take extra steps and create a more robust Employer Profile when they can earn free credits for each step taken.
Key Tasks & Skills: Reviewing candidates takes time that small business owners don’t have. This feature would call out key phrases or skills that Monster identified based on the job title. Users could skim through the candidate profiles more quickly, finding the right fit for their open position.