Contacts
+91 7574959400
Follow us:
Get in Touch
Close

Contacts

Ahmedabad, India

+917574959400

info@theaidivision.com

Call us: +91 7574959400
Follow us:
Get in Touch
Get in Touch

HireBot: Autonomous AI Recruitment Agent

Projects

Brief and project description

Deploying hyper-realistic AI Avatars to automate the entire recruitment lifecycle, from sourcing and ID verification to face-to-face technical interviewing.

HireBot is an end-to-end recruitment ecosystem developed by The AI Division that merges Generative AI with enterprise-grade biometric security. It serves as an autonomous "Gatekeeper," handling the heavy lifting of the hiring funnel. Unlike standard text-based screeners, HireBot conducts face-to-face video interviews using a responsive AI Avatar. It sources candidates, validates their identity to prevent fraud, conducts adaptive technical interviews, and delivers detailed analytical scorecards to hiring managers—reducing time-to-hire from weeks to days.

  • Biometric Identity Gatekeeping: Prevents "proxy interviewing" by verifying candidate IDs against live video feeds before the session starts.

  • Hyper-Realistic Avatar Interface: Conducts natural, verbal interviews with human-like expressions and latency.

  • Adaptive Questioning Engine: Dynamically generates follow-up questions based on the candidate's specific answers, not just a fixed script.

  • Bias-Free Evaluation: Scores candidates purely on technical merit and sentiment, removing unconscious human bias.

  • Automated Logistics: Handles all sourcing, matching, and calendar scheduling without recruiter intervention.

Let's connect

The Challenge

Our client, a rapid-growth Remote-First Technology Company, was battling the dual challenges of scale and security. As they expanded their global engineering team, the traditional hiring process began to fracture.

  • The “Proxy Interview” Epidemic: The client saw a 15% rise in identity fraud, where a skilled individual would take the interview, but a less qualified person would show up for the job. This “bait-and-switch” was costing the company thousands in turnover costs and security risks.

  • Screening Fatigue & Bias: The HR team was overwhelmed by thousands of inbound applications. Recruiters were spending their entire week on repetitive “Level 1” screening calls, asking the same five questions. This repetition led to “interviewer fatigue,” resulting in subjective decision-making and high-quality candidates slipping through the cracks.

  • Process Bottlenecks: Coordinating schedules across four different time zones meant it often took 2 weeks just to book a first-round interview, causing top talent to accept offers elsewhere.

In a remote-first world, trust is the hardest metric to verify. We didn't just need a faster way to interview; we needed a secure way to verify who was on the other side of the screen. HireBot solved both speed and security in one platform.

Director of Talent Acquisition, The AI Division
CargoMind The AI Division

The Solution

We engineered a secure, multi-stage AI Interview Pipeline that functions as a 24/7 autonomous recruiter.

The Workflow Breakdown

  • Phase 1: Smart Sourcing & Scheduling
    The system acts as the initial filter. It ingests resumes and utilizes semantic matching against the Job Description (JD)—looking for contextual experience rather than just keywords. It automatically engages high-match candidates via email and syncs with the team’s calendar to book a video slot instantly.
  • Phase 2: Biometric KYC (The Security Layer)
    This is the core differentiator. When the candidate joins the video link, the AI requests a valid government ID. Using Computer Vision, it verifies the document’s authenticity and performs a real-time 3D face match against the live video feed. If the face doesn’t match the ID, the interview is immediately terminated.
  • Phase 3: The Avatar Interview (The Core)
    A photorealistic AI Avatar conducts the technical screen. Unlike a rigid form, the Avatar listens. It asks adaptive follow-up questions (e.g., “You mentioned using React Context; can you explain a specific drawback you faced with that approach?”). It evaluates technical depth, communication clarity, and English proficiency in real-time.
  • Phase 4: Insight Reporting
    Post-interview, HireBot doesn’t just give a “Pass/Fail.” It generates a comprehensive Hiring Scorecard for the manager. This includes a transcript, a sentiment analysis graph, technical grading, and even “Red Flag” highlights for inconsistent answers.

Technology Stack

We combined Generative AIwith computer vision to create a seamless, secure experience:

  • Avatar Generation: HeyGen API / D-ID (For generating the lip-synced, realistic video avatar).

  • Computer Vision: OpenCV & AWS Rekognition (For Identity Verification and Liveness Detection).

  • LLM Logic: Claude 3.5 Sonnet (Selected for its superior nuance in evaluating human soft skills and technical explanations).

  • Frontend Architecture: Next.js & WebRTC (For low-latency video streaming).

  • Scheduling: Cal.com API (For handling time-zone synchronization).

The Results

HireBot standardized the hiring process, making it faster, fairer, and safer:

  • 100% Elimination of Identity Fraud: The biometric gatekeeping layer completely stopped proxy interview attempts, ensuring only verified candidates entered the pipeline.

  • 60% Reduction in Time-to-Hire: By automating sourcing and L1 screening, the time from “Application” to “Final Round” dropped from 3 weeks to 5 days.

  • Standardized & Fair Evaluation: Every candidate was asked the same core questions in the same professional tone, removing interviewer bias regarding accent, appearance, or background.

Elimination of Identity Fraud
0 %
Reduction in Time-to-Hire
0 %