System Requirements Document (SRD)

1. Introduction

The AI Interviewer Application is designed to streamline the hiring process by automating candidate interviews and evaluations. The system allows HR professionals to create job descriptions, upload relevant documents, invite candidates, and conduct AI-driven interviews. The AI interviewer dynamically generates job-specific questions based on the job description and candidate resume, records the session, and provides an evaluation report. Additionally, HR can adjust weights for technical and soft skills in the evaluation engine per job role, ensuring tailored assessments.

2. System Overview

The AI Interviewer Application is a web-based platform that simplifies recruitment workflows. It consists of three main modules:

1. Admin/HR Panel: For job creation, document upload, candidate invitation, and evaluation weight adjustments.
2. Candidate Interview Flow: A seamless interview experience with an AI bot.
3. Evaluation Engine: Automated scoring and reporting based on interview performance, with customizable weights for technical and soft skills.

The system leverages AI technologies to generate job-specific questions, conduct real-time voice interviews, and evaluate candidates effectively.

3. Functional Requirements as Story Points

• As HR, I should be able to create a job by entering job title, description, skills, experience, and location.
• As HR, I should be able to upload job description documents, company policies, and reference documents.
• As Candidate, I should be able to upload my resume and ID proof (optional).
• As AI, I should be able to parse resumes and extract skills.
• As AI, I should be able to calculate JD vs Resume match percentage.
• As AI, I should be able to flag fake or inconsistent data using basic NLP techniques.
• As HR, I should be able to invite candidates by entering their email addresses.
• As System, I should send secure interview links, auto-signup details, and calendar expiry notifications to candidates.
• As Candidate, I should be able to log in using the secure interview link.
• As Candidate, I should be able to upload my resume and allow mic and camera access.
• As Candidate, I should be able to start the interview by clicking "Start Interview."
• As AI, I should generate job-specific technical, behavioral, and scenario-based questions based on the job description.
• As AI, I should generate project-specific and role-specific questions based on the candidate's resume.
• As AI, I should dynamically adapt follow-up questions based on the candidate's answers.
• As AI, I should conduct real-time voice interviews using speech-to-text and text-to-speech technologies.
• As AI, I should generate a candidate scorecard with scores for technical skills, communication, problem-solving, and JD match percentage.
• As HR, I should be able to adjust weights for technical and soft skills in the evaluation engine per job role.

4. User Personas

Admin/HR

• Responsible for creating job descriptions, uploading documents, inviting candidates, and adjusting evaluation weights.
• Requires tools for document verification, candidate management, and tailored assessments.

Candidate

• Participates in AI-driven interviews.
• Uploads resumes and ID proofs.
• Requires a seamless and secure interview experience.

AI Bot

• Conducts interviews by asking dynamic questions based on job descriptions and resumes.
• Evaluates candidates and generates scorecards.
• Adapts follow-up questions dynamically based on candidate responses.

5. Visuals Colors and Theme

The application will have a professional and modern theme with the following color palette:

• Primary Color: #0052CC (Deep Blue) for headers and buttons.
• Secondary Color: #F4F5F7 (Light Gray) for backgrounds.
• Accent Color: #FF5630 (Red-Orange) for alerts and highlights.
• Text Color: #172B4D (Dark Gray) for readability.

The theme will emphasize simplicity and clarity, ensuring a user-friendly interface.

6. Non-Functional Requirements

• The system should ensure secure data transmission using HTTPS.
• The system should support high availability and scalability to handle multiple interviews simultaneously.
• The system should provide a responsive design for both desktop and mobile devices.
• The system should ensure compliance with data protection regulations (e.g., GDPR).
• The system should have a user-friendly interface with minimal learning curve.
• The system should provide accurate AI-generated questions and evaluations.
• The system should allow HR to adjust evaluation weights without requiring technical expertise.

7. Tech Stack

Frontend

• React for Web

Backend

• Python
• FastAPI

Database

• MySQL or MariaDB (using Alembic for migrations)

AI Models

• GPT 5.2 for user-friendly response generation
• Claude 4.5 Opas for academic or coding work
• Gemini 3 Pro for friendly responses

AI Tools

• Litellm for LLM Routing
• Langchain

Local Orchestration

• Docker
• docker-compose

Server-Side Orchestration

• Kubernetes

Real-Time Voice Interview

• WebRTC
• Whisper / Deepgram
• OpenAI TTS

8. Assumptions and Constraints

• The AI interviewer will generate questions based on job descriptions and resumes provided by HR and candidates.
• The system will rely on AI models for dynamic adaptability during interviews.
• The system will require candidates to have stable internet connections for real-time interviews.
• The system will support English as the primary language for interviews.
• The MVP will focus on core functionalities, with additional features planned for future iterations.
• HR will have the ability to adjust weights for technical and soft skills in the evaluation engine per job role.

9. Glossary

• AI Bot: Artificial Intelligence system conducting interviews and evaluations.
• HR: Human Resources personnel responsible for recruitment.
• JD: Job Description, detailing the requirements for a specific job role.
• Resume Parsing: Extracting relevant information from a candidate's resume.
• NLP: Natural Language Processing, used for analyzing text data.
• WebRTC: Web Real-Time Communication, enabling real-time voice and video communication.
• TTS: Text-to-Speech, converting text into spoken words.
• Scorecard: A report generated by the AI bot evaluating a candidate's performance.
• Evaluation Weights: Customizable parameters for scoring technical and soft skills during candidate assessments.

This updated SRD incorporates the functionality for HR to adjust weights for technical and soft skills in the evaluation engine per job role, ensuring tailored assessments and a balanced approach to candidate evaluation.