AI-Powered Recruiting Copilot
PolarisJobs
Developed an automated interview process for staffing agencies.
PolaisJobs is an AI-Powered Recruiting Copilot that combines AI support, recruiter tools and candidate workflows to simplify hiring for staffing agencies. It aims to leverage AI/ML to support, but not replace, human decision-making in candidate evaluation and placement.
MY IMPACT
Led end-to-end design of third product phase, aligning user needs with product strategy through research, IA, and user flow optimization.
Designed the AI-driven interview flow and Copilot scheduling system, improving recruiter efficiency and reducing time-to-fill by 30%.
Conducted user research to refine workflows, improved handoff efficiency by 30%, and contributed to securing $200K in pre-seed funding from Harvard Innovation Labs and Spark Grants.
Streamlining the Recruitment Process
Allow recruiters to streamline their workflow by scheduling automated interviews, sending confirmations, and editing interview questions in one go.
1
Session Details
2
Email Confirmation
3
Question Editor
4
Confirm
Add a Session
Fill in session details and email information.
*
Session deadline
March 26, 2025
Time
10:00 am
*
Time zone
Pacific Time - Los Angeles
By the end of the day
Name your session
E.g. Interview with John
*
Client name
Technology
*
Job position
Junior Software Engineer
*
Candidate name
John Doe
Email Confirmation
Fill in session details and email information.
Email Confirmation
JohnDoe@gmail.com
*
Subject
E.g Interview Invitation for [Job Title] at [Company Name]
Dear John Doe,
Thank you for applying to the Junior Software Engineer position at Codesphere.
We’re pleased to invite you to complete an asynchronous video interview through PolarisJobs. You can begin the interview at your convenience using the link below:
[Interview Link]
If you have any questions or encounter any issues, please don’t hesitate to reach out. We look forward to learning more about you.
Sincerely,
Back
Next
Question Editor
Edit your interview questions here.
Select or Upload Questions
*
Interview title
Junior Software Engineer Interview
Retake limit per question
3
Session time limit
20
mins
Interview description
This interview evaluates your foundational programming knowledge, problem-solving process, and approach to teamwork. Most questions are short-answer to let you demonstrate clear, concise thinking without needing to write lengthy essays or large programs.
33/500
Interview Questions
1.
2.
3.
4.
5.
6.
Add a question
Can you walk us through your most impactful backend project?
What technologies did you use, and how did you handle scaling?
How have you collaborated with other teams like DevOps or Product?
Do you have leadership or mentorship experience?
Name one programming language or framework you’d like to learn and why.
Describe a recent time you solved a technical problem you didn’t know the answer to at
first. How did you figure it out?
Automated Interview created
You’ve successfully set up the automated interview.
Improving recruiter efficiency and reducing time-to-fill by 30%.
Unified Experience
Instead of spending hours on scheduling and reviews, recruiters can streamline their workflow while maintaining the human element through customizable interview experiences.
Home
Jobs
Interview
Session Dashboard
Recent sessions
Sort by Status
Thurs, March 20
CLIENT NAME - CLIENT JOB POSITION
John Doe - Second Interview (Meeting title)
Thurs, Mar 20, 2025
Expired
CLIENT NAME - JOB POSITION
John Doe - Technical test
Thurs, Mar 20, 2025
Completed
Sunday, March 30
CLIENT NAME - JOB POSITION
Rachel Doe - First Interview
Sun, Mar 30, 2025
In Progress
CLIENT NAME - JOB POSITION
Rachel Doe - Behavioral Question
Sun, Mar 20, 2025
·
4:30 PM
·
2 hours 8 mins
Sent
Overview
Number of Sessions
15
Completed
4
In Progress
10
Expired
1
Awaiting Review
2
Completed Mar 20
John Doe - Technical Test
Completed Mar 22
Annie Doe - First Interview
Question Management
Add a session
Automated
2
Copilot
The Challenge
Staffing agencies have to spend hours on initial candidate screenings and scheduling interviews.
This manual input method was inefficient and delayed placements. Key user problems included:
Scheduling friction
Interview completion inefficiency
Miscommunication around timing, preparation, and feedback
The goal was to streamline interview coordination, reduce manual overhead, to help recruiters improve productivity.
We've built an Automated & Copilot Interview Agent module that bridges gaps between recruiters and candidates by automating core interview workflows.
Phase 3 Goals
Automate Candidate
Screening
Enable recruiters to conduct automated candidate screenings using an AI-powered agent.
Content Analysis
Provide analytical insights into candidate responses.
Improve Efficiency
Significantly reduce the time recruiters spend on initial candidate screening.
Enhance Trust
Build trust with design partners by addressing their immediate screening needs.
Scalability
Design a solution that can evolve to offer more in-depth interview capabilities in the future.
Discovery
To start building the Agent module, first I conducted competitive analysis to understand where to position our product.
After aligning the team with the product visions, I then moved into primary research to understand the Recruiter's stories.
Key Insights from Interview
Collaboration
External collaboration with hiring managers collaboration is the most important
Candidate Re-answer Feature
All participants support candidate re-record option
Recruiters want visibility into number of attempts used
Review Preferences
Transcript and AI summarized notes are deemed the most valuable response formats
Review Efficiency
Recruiters desire tools for automatic flagging of specific keywords/ skills
Structuring User Flow
Drawing from competitive analysis and user research with staffing agencies, I developed user flows for both recruiters and candidates, integrating their feedback.
Recruiter User Flow
Recruiter Priorities
Access real-time interview insights and performance summaries
Review and compare candidate readiness through AI-generated reports
Easily create and manage new interview sessions
Navigate seamlessly between monitoring, reviewing, and planning actions
Candidate User Flow
Candidate Priorities
Provides clear instructions and relevant information to guide candidates through each interview step
Allows technical setup for preparation and practice
Offers flexibility and maintain control during the session
Enables candidates to confidently showcase their skills and potential through a supportive, user-centered flow
Iterations based on Feedback
Through initial testings, I've made iterative improvements on the low-fi structure to translate insights into high-fi prototypes.
Outcomes
Through initial testings, I've made iterative improvements on the low-fi structure to translate insights into high-fi prototypes.
Reflection
What I've Learned
Learned to balance features for AI implementation and feasibility with recruiter needs for building Interview Agent Module
Gained expertise in product strategy and delivering development-ready designs
Gained confidence in articulating ideas and leading responsibilities with ownership
Developed a unified UI style guide, strengthening brand identity and accelerating development efficiency
Areas of Improvement
For future work, I would aim to formally document all technical constraints and design trade-offs made during the process to help the team understand and align on why certain design decisions were made
Conduct more testings on edge cases specifically to identify unusual or failure scenarios
Build a more continuous feedback loop for gathering micro-feedback from live users (even internal users if pre-launch) directly within the prototype or application