Copyright © 2026 Authors retain the copyright of this article. This article 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.

@article{205696,
        author = {Ambika Patil},
        title = {AI-Based Interview Preparation and Skill Assessment Platform},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {1},
        pages = {7351-7355},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=205696},
        abstract = {Technical interviews remain one of the most challenging stages of campus and industry recruitment, primarily because candidates lack a personalised, on-demand way to practice role-specific questions and to objectively gauge their own readiness before a real interview. This paper presents the design and implementation of an AI-Based Interview Preparation and Skill Assessment Platform built on the MERN stack (MongoDB, Express, React, Node.js) and integrated with Google's Gemini large language model through the @google/genai SDK. The system allows an authenticated user to create a preparation session by specifying a target job role, years of experience, and topics to focus on; the backend then prompts the Gemini model to generate a structured set of question-answer pairs that double as a self-assessment instrument, repairs and validates the returned JSON, and persists the questions against the session in MongoDB. A secondary endpoint allows the user to request an on-demand, in-depth concept explanation for any saved question; while pinning and note-taking features let users track which skill areas need further revision. The platform uses JSON Web Tokens for stateless authentication and a React/Vite single-page front end for session management, dashboarding, and an interactive question-and-answer interface. We describe the system architecture, the prompt-engineering and JSON-repair strategy used to make LLM output reliable, the data model, and the REST API surface, and we discuss the practical challenges of integrating a generative AI service into a production-style web application. The resulting system demonstrates that a lightweight MERN application augmented with a generative AI backend can produce a scalable, low-cost platform for both personalised interview preparation and ongoing self-assessment of technical skill gaps.},
        keywords = {Artificial intelligence, Gemini API, interview preparation, JSON Web Token authentication, large language models, MERN stack, MongoDB, prompt engineering, React, skill assessment, web applicationI.},
        month = {June},
        }