028 - Ongoing projects – Making them market-ready with AI-based enhancements

2027 · 2027 Competition

School: School of Computer and Information Sciences
Category: Corporate SponsoredPrimary

Project Overview

One Liner: Ongoing projects – Making them market-ready with AI-based enhancements

Abstract

FMR (Face My Resume) – It’s a tool that makes the Hiring manager’s life easier by reducing the recruitment time and cost by upto 50%, working through the real-time video resume.

Hiring teams often juggle candidate information, video assessments, interview feedback, and scheduling across separate tools, which makes the process slow, inconsistent, and hard to track. Within that process, reviewing candidates is the biggest bottleneck: reviewers must watch each video answer, take notes, rate it, and compare candidates manually.

Face My Resume (FMR) addresses this by centralizing the hiring workflow in one platform. Recruiters create positions, attach questions, and invite candidates. Candidates upload a traditional resume and record short video answers that show their technical and soft skills. Reviewers then rate the videos on customizable criteria and move candidates to Select, Reject, or Maybe, and managers schedule interviews. FMR is live in production at facemyresume.com and aims to cut recruitment time and cost by up to 50%. However, it has no AI features yet. Video review and rating are fully manual, question sets are attached to each job by hand, expired candidate availability has no recovery path, and approved jobs are published to the website manually.

This project will make FMR market-ready by adding AI-based enhancements on top of the existing workflow. The sponsor has identified six: AI question filtering and auto-rating, expired availability rescheduling, job-based question assignment, auto job publication, video assessment to text and rating, and body language analysis. The team will focus on the video assessment pipeline, where a video answer is converted to a transcript, summarized, and given a rubric-based draft rating that a reviewer confirms or edits. A second track covers role-to-question mapping using retrieval, and the body language and workflow features are treated as research or stretch goals. The final scope will be confirmed with the sponsor at kickoff.

All AI work follows the sponsor’s responsible-AI guardrails. AI output only assists reviewers and never makes a hiring decision. Ratings must show the rubric criteria and transcript evidence behind them, and body language signals are reported as observations but kept out of the candidate score. The team will also test for bias, protect candidate privacy by using only anonymized or synthetic data, and note regulatory implications such as NYC Local Law 144, the Illinois AI Video Interview Act, and the EU AI Act.

No video available.

Screenshots

0 image(s)

No screenshots uploaded yet.

Team Members

Renfei Song
Lead
Andy Pham
Xin Ni
Velina Wang
Hanjun Xu
Pankti Patel

Advisors

Kai Zhang
Kai Zhang

Stakeholders

Sandra Dixit
Al Limaye