Shortlist.Ai

Shortlist.ai uses AI to filter, interview, and rank job applicants—so recruiters find top talent fast. We replace keyword spam with vector matching and voice interviews for smarter and faster hiring.

Comment

SLIDESHOW: https://www.canva.com/design/DAGrDlCVX2E/fpRBLP3lqfQkMus4zkTxRA/edit?utm_content=DAGrDlCVX2E&utm_campaign=designshare&utm_medium=link2&utm_source=sharebutton

Inspiration

We spoke with recruiters at Zebra Technologies, a Fortune 500 company, who told us they had 800 applicants for a single role-but only reviewed 30. Not because they wanted to, but because they didn’t have both time or an effective way to filter them. We realized that today’s hiring systems aren’t broken because of a lack of applicants—they’re broken because recruiters don’t have time or tools to find the right ones. That’s where Shortlist.ai began.

What it does

Shortlist.ai is a recruitment assistant that uses AI to filter, interview, and rank applicants. It transforms job descriptions and resumes into vector embeddings to semantically match candidates, then conducts voice-based interviews using LLM-generated questions and dynamic follow-ups. We also run automated background checks before shortlisting to ensure candidates meet company standards. Finally, applicants are scored from 1 to 10—so recruiters get a short, smart list without wasting time.

How we built it

flask servers communicate with each other via ngrok tunneling port. twilio api handles ai agent phone screen. vector calculations performed by scikit and generated using openai ada 003 embedding model.

Challenges we ran into

  • connecting the phone twilio service to an ai agent
  • connecting the twilio service and phone screen to the app.. we had to make 2 new servers instead and share data between them via supabase uploads.

Accomplishments that we're proud of

Built a fully functional vector-matching + AI interview pipeline Conducted actual voice calls with LLM-powered follow-ups Got validation from Zebra Technologies, who are actively interested in piloting the tool Developed a visual recruiter dashboard to monitor candidate scores and resumes

What we learned

Vector-based ranking systems can dramatically improve hiring fairness over traditional hiring methods like ATS Recruiters want tools that enhance—not replace—their process

What's next for Shortlist.Ai

Our next step is running a pilot with Zebra Technologies, a Fortune 500 company that received 800+ applicants for a single role but only had time to review 30. With Shortlist.ai, we’re helping them filter, interview, and rank every candidate—automatically. This pilot will validate our system in a real-world hiring environment and lay the groundwork for scaling to similar-sized companies facing the same hiring bottlenecks.

Built With

Share this project:

Updates