AI-Driven Voice Based Personalized Mock Interview System

Abstract

Effective interview preparation requires personalized guidance based on a candidate’s skills, experience, resume, and target role. Existing platforms often rely on generic interview questions and provide limited performance feedback. This paper proposes an AI-Driven Voice-Based Mock Interview and Career Preparation Platform that integrates personalized technical and managerial mock interviews with ATS-based resume analysis, resume building, and career roadmap generation. The system analyses candidate information to generate role-specific questions, evaluate responses, identify skill gaps, and provide performance scores and actionable feedback. It also generates personalized learning paths and relevant resources based on identified skill deficiencies. By integrating these capabilities into a unified platform, the proposed approach aims to enhance interview readiness, career preparation, and overall job readiness through personalized and AI-driven guidance.

Introduction

The text presents an AI-based Career Counselling and Job-Readiness Platform designed to provide personalized career guidance and help candidates prepare for employment. The system addresses limitations of traditional career counselling, which often provides generic advice without considering an individual's skills, resume, education, career interests, and target job role.

The platform combines Artificial Intelligence (AI), Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval-Augmented Generation (RAG) to provide several interconnected career services.

Main Features

• Personalized Career Analysis: Examines a candidate's education, skills, resume, interests, experience, and target role to determine their current career status.

• AI Mock Interviews: Generates technical and managerial questions specific to the candidate and target job, then evaluates answers and provides scores, strengths, weaknesses, and improvement suggestions.

• AI Resume Scoring: Evaluates resumes using ATS-related factors such as keywords, structure, skills, and relevance to the target role.

• AI Resume Builder: Creates structured and ATS-friendly resumes using the candidate's education, projects, skills, and experience.

• Skill-Gap Identification: Compares existing candidate skills with the requirements of the desired job role to identify missing competencies.

• Personalized Career Roadmap: Creates a learning plan containing recommended technologies, resources, and practical projects based on identified skill gaps.

• Integrated Job-Readiness Assessment: Combines resume quality, interview performance, and skill assessment to provide a broader view of a candidate's preparation.

• Data-Driven Guidance: Uses candidate information and AI-generated assessments to provide personalized recommendations.

Literature Review

Existing research demonstrates the usefulness of AI for resume analysis, ATS evaluation, interview-question generation, conversational interviews, behavioral assessment, and emotion recognition. However, these systems generally focus on individual tasks rather than providing an integrated career-development platform.

The identified gaps include limited support for:

• Managerial interview assessment.

• Resume generation and ATS optimization.

• Comprehensive skill-gap analysis.

• Personalized learning resources.

• Career-roadmap generation.

• Combining resume, interview, and skill assessments.

• Fully personalized recommendations based on the candidate's target role.

The proposed system attempts to address these gaps by integrating multiple career-preparation functions into one platform.

Proposed Methodology

The system follows an integrated workflow:

Candidate Data → Resume/Profile Analysis → AI Assessment → Skill-Gap Detection → Personalized Recommendations → Career Roadmap

Candidate information such as resumes, education, experience, skills, and target job roles is first collected and processed. The AI modules then analyze this information to generate interview questions, evaluate responses, score resumes, identify skill gaps, and recommend learning paths.

The modules are interconnected. For example, resume analysis and mock-interview performance can be used for skill-gap identification, and those identified gaps can then be used to generate a personalized career roadmap.

System Architecture

The platform uses a microservices-based architecture, which separates major functions into independent services and improves scalability and maintainability.

The major components are:

1. Client Layer: A React.js frontend for registration, profile management, resume uploads, interviews, resume analysis, and career-roadmap access.

2. API Gateway: Routes requests between the frontend and backend services while handling common communication and validation tasks.

3. Authentication/User Service: Manages registration, login, profiles, and role-based access using technologies such as JWT or Firebase Authentication.

4. Resume Service: Handles resume uploading, parsing, information extraction, ATS scoring, and resume generation.

5. Interview Service: Generates personalized technical and managerial interview questions and evaluates candidate responses.

6. AI/LLM Service: Provides the core intelligence using LLMs, LangChain, LangGraph, and potentially RAG for contextual information retrieval.

7. Skill-Gap and Roadmap Service: Compares candidate abilities with job requirements and generates customized learning plans.

8. Data Storage Layer: Uses MongoDB for application data and a vector database such as Qdrant for semantic-search embeddings.

9. Infrastructure Layer: Uses REST APIs, Docker, AWS, and potentially Redis for communication, deployment, scalability, and caching.

Conclusion

The proposed AI-Driven Voice-Based Mock Interview Platform provides a unified and personalized approach to interview and career preparation. The system integrates resume analysis, ATS scoring, personalized question generation, voice-based mock interviews, response evaluation, skill-gap detection, feedback, and career roadmap generation. By using the candidate’s resume, skills, experience, and target role, the platform can provide a more relevant interview experience than generic preparation systems. The evaluation and feedback process helps candidates understand their strengths and weaknesses and identify areas requiring improvement. Overall, the proposed system aims to make interview preparation more personalized, accessible, data-driven, and effective, thereby improving candidate job readiness and employability. Future development can focus on improving the intelligence and personalization of the platform. The system can be extended with multimodal analysis by combining voice, text, and visual interview signals for more comprehensive performance evaluation. Advanced models can be used to improve confidence and communication analysis, adaptive question generation, and response evaluation. Future versions can also include real-time interview analytics, multilingual interview support, improved skill-gap prediction, and integration with job-market requirements. A larger dataset and real-world user testing can further improve the reliability and accuracy of the proposed system.

References

[1] Mrs. Manjula Jayamma, Dr. C. Venkataiah, G. Kowsalya “ Gen-AI Interview Simulator & Resume Advisor for Top IT Firms ” 2026 4TH International Conference on Sustainable Computing and Smart Systems (ICSCSS-2026) IEEE Xplore Part Number: CFP26DJ3-ART; ISBN: 979-8-3315-8307-1 [2] Karthik Yadav, Chaitanya Agarwal, Sudha Senthikumar “AI-Powered Interview Platform” 2026 5th OPJU International Technology Conference (OTCON) on Smart Computing for Innovation and Advancement in Industry 5.0 | 979-8-3315-6149-9/26/$31.00 ©2026 IEEE | DOI: 10.1109/OTCON69332.2026.11629826 [3] Pugazhendhi S, Mohammad P, Dr. Sherly Pushpa Annabel “AI-based Real Time Mock Interview System using Generative and Conversational Agents” Proceedings of the 4th International Conference on Inventive Computing and Informatics (ICICI-2026) IEEE Xplore Part Number: CFP26L34-ART; ISBN: 979-8-3315-5730-0 [4] Ganesh Kumar Yadav, Mohd Saim Rafi, Pranjal Vijayvargiya “Real Time AI-Based Virtual Interview Panel” 2026 International Conference on Sustainable Engineering and Technology Innovations (ICSETI) [5] Xiaohui Liang “Resume Automatic Analysis and Matching Technology” 2025 International Conference on Digital Analysis and Processing, Intelligent Computation (DAPIC) [6] Nguyen, T. T. H., Nguyen, T. D. Q., Cao, H. L., Tran, T. C. T., Truong, T. C. M., & Cao, H. (2025). SimInterview: A multilingual LLM-based simulated interview training system. (ICEFM 2025) . https://arxiv.org/abs/ 2508.11873 [7] Artiran, S., Shukla, V., & Ahmed, R. (2024). Emotion-sensitive AI interviewers with adaptive multimodal responses. Journal of Intelligent Human-Computer Interaction, 48(3), 210–225. [8] Luo, J., Wei, X., & Zhang, Q. (2024). Stress detection n AI interview systems using facial and speech cues. IEEE Access, 12, 45789–45802. [9] Geng, L., Patel, A., & Bose, R. (2024). Virtual reality-based mock interview training with biofeedback integration. Computers & Education Technologies, 45(4), 301–319. [10] Migovich, D., Kim, J., & Singh, P. (2024). Human-AI hybrid panels for behavioural interview simulation. ACM Transactions on Interactive Intelligent Systems, 34(1), 98–113. [11] Singh, P., Kumar, R., & Desai, A. (2024). Bridging psychological modelling and AI for behavioural training. Cognitive Computing and Systems, 17(1), 89–104. [12] Patel, N., Sharma, K., & Banerjee, R. (2023). Virtual Reality Exposure Therapy and AI integration for anxiety management. ACM Transactions on Human-Computer Interaction, 29(5), 102–115. [13] Rahman, T., Singh, D., & Bose, R. (2023). Advances in multimodal affective computing and emotion recognition. Journal of Artificial Intelligence Research, 76(2), 310–329. [14] Naim, I., De La Torre, F., & Morency, L.-P. (2021). Automated scoring of candidate interviews using multimodal speech and linguistic analysis. IEEE Transactions on Affective Computing, 12(4), 670– 684. [15] Naim, I., Tanveer, M. I., Gildea, D., & Hoque, M. E. (2015, May). Automated prediction and analysis of job interview performance: The role of what you say and how you say it. In 2015 11th IEEE international conference and workshops on automatic face and gesture recognition (FG) (Vol. 1, pp. 1-6). IEEE. [16] Poria, S., Cambria, E., Bajpai, R., & Hussain, A. (2017). A review of affective computing: From unimodal analysis to multimodal fusion. Information fusion, 37, 98-125.

Copyright

Copyright © 2026 Prof. Kalyani Dahikar, Ritesh Pundkar, Prasad Rathod, Pranav Lande, Tejas Patre. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.