AI Resume Studio: An Intelligent Full-Stack Career Platform Leveraging Large Language Models for Resume Analysis, Skill Gap Detection, and Interview Preparation

Abstract

The proliferation of artificial intelligence across professional domains has ushered in a new era of intelligent career tools that fundamentally transform how candidates prepare for and navigate the job market. This paper presents AI Resume Studio, a comprehensive, full-stack, AI-powered career platform engineered using Next.js 16, Supabase, and Google Gemini AI, augmented by OpenAI language models. The system integrates a drag-and-drop resume builder supporting professional templates with photo capabilities, applicant tracking system (ATS)-based resume scoring and optimization, automated skill gap detection with AI-driven suggestions, an embedded interview preparation module with mock tests, and a semantically filtered job board powered by real-time Supabase queries. The platform employs a multi-tiered architecture comprising a Next.js frontend, a FastAPI microservice layer, and a Supabase PostgreSQL database, enabling secure role-based authentication and scalable data management. Empirical evaluation demonstrates a 28% improvement in ATS scoring accuracy over baseline systems, a 75% reduction in resume parsing latency, and a 22% measurable improvement in candidate interview performance as reported through structured user studies. The system addresses critical shortcomings in existing career tools by unifying disparate functionalities into a single cohesive, AI-augmented workflow, thereby reducing candidate preparation time and significantly improving employment readiness outcomes.

Introduction

The text presents AI Resume Studio, an integrated AI-powered career platform designed to solve the problem of resume screening in highly competitive job markets where most applications are filtered by ATS systems before human review.

The system combines multiple career tools into a single platform, including resume building, ATS optimization, skill gap analysis, AI-generated suggestions, interview preparation, and a semantic job board. It is built using a modern architecture with Next.js for the frontend, FastAPI microservices for AI processing, Supabase for authentication and data storage, and Google Gemini for language-model-based analysis.

Key features include drag-and-drop resume creation, automated ATS scoring with feedback, personalized skill gap detection with learning recommendations, and AI-generated mock interview questions. The system also provides job matching based on semantic similarity between user profiles and job descriptions.

An experimental evaluation with 120 participants shows strong improvements over existing tools, including:

• +28% ATS accuracy improvement
• 75% reduction in processing time
• 87% skill gap detection precision
• 4.7/5 user satisfaction
• 22% improvement in interview performance

The study concludes that integrating all career preparation stages into a single AI-driven platform significantly improves both resume quality and candidate performance, outperforming fragmented existing tools like resume builders, ATS scanners, and interview platforms.

Conclusion

This paper presented AI Resume Studio, a full-stack intelligent career platform that integrates resume construction, ATS optimization, skill gap detection, AI-powered remediation, interview preparation, and job discovery within a unified, authenticated user experience. The system leverages the capabilities of Google Gemini and OpenAI language models, deployed through a scalable microservices architecture built on Next.js 16, FastAPI, and Supabase, to deliver contextually intelligent career guidance at scale. Empirical evaluation across 120 participants over eight weeks demonstrated statistically significant improvements in ATS scoring accuracy (89%), skill gap detection precision (87%), user satisfaction (4.7/5 CSAT), and interview performance (22% improvement). These results establish AI Resume Studio as a significant advancement over existing career tools and validate the design philosophy of integrated, AI-augmented career intelligence. The platform\'s open architecture facilitates further research and extension, and the modular microservices design supports independent scaling of AI components as demand grows.

References

[1] J. Schmitt, M. Entrup, and A. Nürnberger, \"Resume information extraction with cascaded hybrid model,\" in Proc. ACL Workshop NLP Human Resources, Melbourne, 2018, pp. 1–9. [2] L. Guo and X. Chen, \"A neural approach to resume-job matching using contextual embeddings,\" in Proc. IEEE ICTAI, Baltimore, MD, USA, 2020, pp. 1012–1018. [3] K. Kenthapadi, B. Le, and G. Venkataraman, \"Personalized job recommendation system at LinkedIn,\" in Proc. ACM RecSys, Como, Italy, 2017, pp. 346–354. [4] S. Roy, A. Patel, and R. Mehta, \"Neural ATS simulation for resume screening,\" in Proc. IEEE BigData, Orlando, FL, USA, 2022, pp. 2341–2348. [5] Y. Zhang, J. Wang, and H. Liu, \"AI-driven mock interview system with GPT-3 for technical role preparation,\" in Proc. ACM IUI, Sydney, Australia, 2023, pp. 89–97. [6] A. Vaswani et al., \"Attention is all you need,\" in Advances in Neural Information Processing Systems, vol. 30, 2017, pp. 5998–6008. [7] T. Brown et al., \"Language models are few-shot learners,\" in Advances in Neural Information Processing Systems, vol. 33, 2020, pp. 1877–1901. [8] Google DeepMind, \"Gemini: A family of highly capable multimodal models,\" arXiv preprint arXiv:2312.11805, 2023. [9] P. Verma and S. Singh, \"Skill gap analysis using NLP for engineering graduates,\" Int. J. Eng. Res. Technol., vol. 10, no. 3, pp. 421–428, 2021. [10] M. Naous, A. Fares, and A. Baly, \"Resume scoring and recommendation: A systematic review,\" ACM Comput. Surv., vol. 55, no. 14, pp. 1–34, 2023. [11] Supabase Inc., \"Supabase: The open source Firebase alternative,\" Documentation, 2024. [Online]. Available: https://supabase.com/docs [12] Vercel Inc., \"Next.js 16 documentation,\" 2024. [Online]. Available: https://nextjs.org/docs [13] S. Tirumala et al., \"An ATS-aware resume optimization framework using transformer models,\" in Proc. EMNLP Industry Track, Singapore, 2023, pp. 543–551. [14] R. Anderson and K. Palmer, \"Evaluating AI feedback systems for career coaching applications,\" J. Career Dev., vol. 49, no. 2, pp. 118–133, 2022. [15] W. Liu, P. He, and W. Chen, \"ResumeNet: A learning-based framework for resume quality assessment,\" in Proc. CIKM, Virtual Event, 2021, pp. 3148–3152.

Copyright

Copyright © 2026 Bodapati Harsha Vardhan, Ammisetti Pardha Saradhi Naidu, Bogem Prasad, Pavithra Natarajan. 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.