Recuria is a smart web application designed to revolutionize the traditional hiring process using a modular, multi-agent AI system. It simplifies resume screening, automates interviews, performs emotion analysis, and sends selection/rejection emails — all in one platform.
Recruitment often involves sifting through hundreds of resumes, conducting repetitive interviews, and manually handling candidate communication. Recuria solves this by acting as an intelligent hiring assistant that helps HR teams make faster, smarter, and bias-free hiring decisions.
- Recruiter enters a Job Description
- Uploads multiple CVs
- AI scores and ranks CVs based on relevance
- AI agent places calls using pre-fed or AI-generated questions
- Transcribes and evaluates candidate responses
- Ranks candidates based on verbal answers
- In-app video interview with transcription + emotion & tone analysis
- Generates a detailed candidate interaction report
- Recruiter reviews all stages + cumulative scores
- AI drafts personalized email templates
- Sends selection/rejection emails
Layer | Technology Used |
---|---|
Frontend | ReactJS, Tailwind CSS, Next.js |
Backend | Python (FastAPI / Flask) |
NLP & Scoring | OpenAI GPT-4, LangChain, Sentence-BERT |
Speech & Voice | Twilio, Google STT, Whisper, ElevenLabs |
Emotion Analysis | RAVDESS-trained models, OpenFace, BERT |
Database | MongoDB, PostgreSQL |
SMTP, SendGrid, Nodemailer | |
Deployment | Docker, Render, Vercel, GitHub Actions |
Aspect | Feasibility Level | Notes |
---|---|---|
Technical | ✅ High | All APIs and models are production-ready or open-source |
Operational | ✅ Medium-High | Seamlessly fits into existing HR workflows |
Financial | ✅ Moderate | MVP can run on free-tier services and low-cost APIs |
Scalability | ✅ High | Microservice-based + Dockerized for cloud-native deployments |
User Acceptance | ✅ High | Saves recruiter time, improves hiring quality, and reduces human bias |
- Calling the candidate with its given phone no. in resume and ask him/her questions regarding job and evaluating them
- Making an video call to candidate and evaluating them based on tone and emotion analysis while answering the HR's questions
- Publishing the final report to the user containing the final of the candidate ( resume + call interview + video call interview )
- Sending e-mails to the selected candidates
- Clone the repo
git clone https://github.com/your-username/recuria.git cd recuria
The project is in development stage