# Project export: STIA AI-powered interviews

This document was generated by HackStack to give an AI agent context about a hackathon project. Sections are labeled with their provenance; content marked as truncated was cut to keep this document small.

## Project metadata

- Hackathon: Cal Hacks 12.0
- Tagline: Save your hiring team time and discover stronger candidates with our AI-powered interview platform. We automate behavioral interviews using questions tailored to job needs and specific resumes.
- Devpost: https://devpost.com/software/stia-ai-powered-interviews
- GitHub: not linked
- Demo: http://10.72.1.237:3001/
- Video: https://www.youtube.com/embed/2iYau89Sr0s?enablejsapi=1&hl=en_US&rel=0&start=&version=3&wmode=transparent
- Team: contributor stats unavailable

## Devpost submission (written by the team)

### Inspiration

We found that employers often spend countless hours screening candidates through repetitive interviews and resume reviews. At the same time, many talented applicants never get the chance to showcase their skills due to limited recruiter availability. We wanted to build a system that bridges this gap using AI.

### What it does

STIA is an AI-powered interview agent that automates the initial screening process. It analyzes resumes, conducts online interviews, and evaluates candidate responses using NLP and sentiment analysis.(uses pro and an anti hire agent along with a judgment agent) Based on the interview performance, it decides whether the candidate should move forward in the hiring process — saving time for recruiters while giving more applicants a fair opportunity.

### How we built it

We built STIA using a modern AI-driven tech stack designed for scalability and real-time interaction. The system is powered by Fetch.ai for orchestrating intelligent workflows, LiveKit for hosting live audio/video interviews, and a custom AI voice agent that serves as the interviewer. On the development side, we used Python for backend logic and AI integration, Next.js for the frontend interface, and Supabase for secure user authentication and data management. Together, these components create a seamless experience where candidates can join interviews, interact with the AI agent, and receive dynamic, personalized questions in real time.

### Challenges we ran into

Some APIs didn’t function as expected or had compatibility issues with our setup. • Debugging the workflow logic between the AI agent, Fetch.ai, and LiveKit was time-consuming. • The voice agent occasionally misinterpreted or failed to store interview responses correctly. • Data feedback loops between components caused inconsistencies during testing. • Managing real-time synchronization between the frontend and backend introduced latency issues.

### Accomplishments we're proud of

We managed to get a working prototype that connects the AI interviewer, workflow engine, and meeting system together. The agent can take a resume, conduct a basic interview, and store results in our database. It’s not perfect, but it proves the core idea works and can be scaled further with refinement.

### What we learned

We learned how to integrate multiple AI and communication systems into a single workflow. Handling real-time data transfer between APIs taught us a lot about debugging and synchronization. We also realized how important it is to plan data flow early, since even small errors in feedback loops can break the whole pipeline. Overall, it gave us hands-on experience with building and connecting AI-driven systems end-to-end.

### What's next

We plan to extend STIA beyond hiring and into internal workforce analysis. The same AI framework can be adapted for performance reviews, identifying high-performing employees, and providing managers with structured insights. Over time, it could help organizations understand skill gaps, recommend training, and make fairer promotion decisions — all while keeping human oversight in the loop.

## README (from the GitHub repository)

No README available.

## Detected evidence (automated analysis)

No repository was indexed for this project. Claimed technologies below could not be checked against code.
- JavaScript (language) — claimed on Devpost, not found in the code
- Next.js (technology) — claimed on Devpost, not found in the code
- PostgreSQL (technology) — claimed on Devpost, not found in the code
- Python (language) — claimed on Devpost, not found in the code
- Supabase (technology) — claimed on Devpost, not found in the code

## Codebase structure

No repository index available.

## Key source files

No repository index available; no source files included.