International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 12 Issue: 06 | Jun 2025 www.irjet.net p-ISSN: 2395-0072
Rezumix: An AI-Powered Career Assistant with Resume Enhancer, Personality Insights, Career Guidance & Mock Interviews
Devansh Agrawal
B. Tech CSE, 4th Year
The ICFAI University, Raipur
Email: agrawaldevansh27@gmail.com
LinkedIn: https://www.linkedin.com/in/thedevanshagrawal
Abstract - In today’s competitive job market, simply having a resume isn't enough you also need to know how to present your strengths, identify your ideal career path, and confidently handle interviews. Rezumix is a modular, AI-powered platform that helps users across all these touchpoints. Built using Next.js (App Router), Tailwind CSS, MongoDB, and integrated with Gemini AI, it offers four independent tools: a Resume Analyzer and Enhancer, Personality Predictor, Career Recommendation and Mock Interview Simulator.
Unlike traditional systems, Rezumix focuses on personalization using LLMs, delivering context-aware feedback, suggestions, and recommendations. This paper outlines the motivation behind Rezumix, the technical framework, its AI integration, and a small-scale user study conducted with 40 users. We also compare its features with existing tools like LinkedIn AI and VMock to highlight its uniqueness.
Key Words: AI Resume, Career Guidance, Mock Interview, Personality Prediction, LLMs, Gemini, Google, ChatGPT, OpenAI
1.INTRODUCTION
Navigating career development today requires more than just technical skills. From building the right resume to choosing a career path and preparing for interviews, candidates face challenges that are often ignored by educationalsystems.
Existing tools like LinkedIn’s AI resume assistant, VMock, and HireVue offer some support, but many are limited in customization, accessibility, or depth of feedback. This inspiredthecreationofRezumix,anAI-poweredplatform that offers standalone tools focused on specific stages of careerpreparation.
1.1 Objective & Research Framework
ThispaperpresentsatechnicalimplementationofRezumix whileaimingto:
-DemonstratethefeasibilityofintegratingLLMsintocareer tools
-Evaluateuserfeedbackonfourcoremodules
-Comparefeatureswithexistingtools
-Outlinefuturepotentialandchallenges.
Whilethisisnotatraditionalexperimentalstudy,weoffera proof-of-conceptimplementation,earlyuserresponses,and positioningwithincurrentmarkettools.
2. Tech Stack and System Overview
Rezumixisbuiltentirelyusingmodern,developer-friendly tools:
Frontend: Next.js (App Router), Tailwind CSS, Framer Motion
Backend/APIRoutes:Next.jsServerActions
AILayer:GeminiAPI(Google'spowerfulLLM)
Database:MongoDB
Other Tools: HTML-to-text parsing, PDF processing, React Charts
SystemFlow:
Frontend(Next.js+Tailwind+FramerMotion)
AILayer(GeminiAPI)
Backend(APIRoutes,MongoDB)
Each feature has its own endpoint and logic cleanly separated.
3. Features (Independent Modules)
3.1 Resume Analyzer & Enhancer
Users can upload their resume. The system parses the content,checksforweakpoints,andthenusesGeminiAIto provide personalized suggestions based on industry standards.
Features: ATS check, formatting feedback, keyword enrichment
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 12 Issue: 06 | Jun 2025 www.irjet.net p-ISSN: 2395-0072
FrontendPerks:Real-timesuggestions,animationonresult cards
Unlike LinkedIn’s resume feedback tool which focuses mostlyonATSscoring,Rezumixgivestailoredsuggestions usingGemini’scontext-basedLLMoutputs.
3.2 Personality Prediction System
Usersanswerquickquestions.GeminiAIreadsthecontent, interpretspsychologicaltraits,andgivesfeedbacklike:
“Youappearhighlyopentonewexperiences,whichisgreat forrolesindesignorR&D.”
“You show strong conscientiousness a good fit for structuredtasks.”
Thesystemdoesn'taskforsensitivedata.Itonlyusesuser inputtogeneratemeaningfulinsights.
3.3 Career Recommendation Engine
Notsurewhatjobsuitsyourvibe?Thisenginelooksatyour choseninterests,andrecommends:
-Jobroles
-Skillpaths
Unlikeotherplatforms,thisonedoesn’toverwhelmyou it gives5-10clearideasbasedonyou.
4. User Feedback and Basic Metrics
We ran a beta test with 42 users (30 students, 12 job seekers).Eachparticipantused2–4featuresandcompleted ashortsurvey.
5. Future Scope
-Voiceinterviewsimulatorwithtoneanalysis
-LinkedInoptimizerbasedonyourresume
-Conversationmemorysoitremembersyourgoals
-AImentorshipandcareertrackingdashboards
Future academic work can involve a formal comparative study and usability testing with statistical methods and a largersamplesize.
6. Conclusion
Rezumixisnotasingle-purposetool it’samulti-feature assistantforstudentsandprofessionals.Eachmoduleisbuilt independently, and users can access only what they need. With furtheriterations and researchcollaborations,itcan grow into a robust platform blending personalization, AI, andcareerintelligence.
REFERENCES
[1] GoogleDeepMind,GeminiAPIDocumentation,2024.
[2] LinkedIn AI Resume Assistant, LinkedIn Corporation, 2023.
[3] VMockInc.,ResumeFeedbackandCareerPlatform,
[4] HireVue,AI-PoweredInterviewSystem,
[5] MBTI Foundation, Applications of Personality AssessmentinCareerGuidance,2020.
[6] OpenAI,LargeLanguageModelCapabilities,2024.
ResumeAnalyzer &Enhancer 4.6 95% Clean, fast, made my resume jobready
Career Recommendation 4.4 83% Suggested paths I didn’tthinkof.
Personality Prediction 4.3 78% Fun and insightful. Feltaccurate
Devansh Agrawal is a 4th year B. Tech CSE student at The ICFAI University, Raipur. He specializes in full-stack development, AI integrations, and real-world applicationsystems.Rezumixishis original solo project developed usingtheMERNstack,andGemini API. Feature
MockInterview 4.7 88% Super realistic. Boosted my confidence
Note:Thiswasaqualitativepilotstudy.Nostatisticaltesting was applied, but the feedback showed strong user satisfaction.
[7] CareerBuilder, ATS and Resume Insights, CareerBuilder Survey,2023.
BIOGRAPHIES
