Issuu

SmartCodeHub

Mrs.G.Swetha1 , E.Joshitha2 , B.Ashritha3 , A.Sai Rakshith4

1 Assistant Professor, Department of Computer Science and Engineering 2,3,4 B.Tech Students, Department of Computer Science and Engineering

Teegala Krishna Reddy Engineering College , Telangana, India ***

Abstract - In the modern software development environment, efficient project management and effective debugging tools are essential for improving developer productivity and code quality. Traditional code repositories mainly provide storage and sharing capabilities but lack intelligentsupport for detectingandcorrectingprogramming errors. This project proposes SmartCodeHub, a unified webbased platform that combines project collaboration with AIbasedcodecorrection. Thesystemallows userstoregisterand upload their software projects after receiving approval from an administrator. Each user can manage their own projects while also viewing and downloading projects shared by other users. The platform ensures secure access control by allowing only project owners to modify or delete their uploaded files, while administrators can monitor and manage all users and projects within the system.

In addition to project management features, SmartCodeHub integrates an AI-powered Code Assistant built using the Gemini 2.5 Flash model through the LangChain framework. This intelligent module enables users to submit code containingerrors, automaticallyanalyzethecode,andreceive corrected versions along with clear explanations. The integration of artificial intelligence transforms the platform into a smart development environment that supports both collaboration and learning. The proposed system improves development efficiency, simplifies debugging, and promotes knowledge sharing among developers, making it a valuable tool for students, programmers, and software development teams.

Key Words: SmartCodeHub, Code Repository, AI Code Assistant, Gemini AI, LangChain, Django, Code Debugging, Project Collaboration, Software Development, Web Application.

1.INTRODUCTION

Therapidgrowthofsoftwaredevelopmentandcollaborative programminghasincreasedthedemandforplatformsthat support efficient project management, code sharing, and debugging.Modernsoftwareprojectsofteninvolvemultiple developers working together, which requires reliable repositoriestostore,manage,anddistributecodeefficiently. Traditionalcodehostingplatformsmainlyfocusonproject storageandversioncontrolbutprovidelimitedsupportfor intelligentdebuggingandautomatederrorcorrection.Asa result, developers frequently rely on external tools and

documentation to identify and fix programming errors, whichcanslow down the development processand affect productivity[1],[2].

Withtheadvancementofartificialintelligenceandmachine learning technologies, intelligent systems can now assist developers in analyzing and improving their code. Recent research shows that large language models and deep learning techniques are capable of understanding programming languages, detecting errors, and suggesting improvements automatically [3], [4]. These technologies have opened new opportunities for integrating AI-driven code assistance into development platforms, enabling developerstoreceiveinstantfeedbackandcorrections.

To address these challenges, this project proposes SmartCodeHub, a unified platform that integrates project repository management with an AI-based code correction system.Theplatformallowsusersto upload,manage,and sharetheirprojectswhilealsoreceivingintelligentsupport fordebuggingandimprovingtheircode.

1.1 Code Repository Platforms

Coderepositorysystemsplayavitalroleinmodernsoftware developmentbyprovidingacentralizedenvironmentwhere developerscanstoreandsharetheircode.Platformssuchas Git-based repositories enable users to collaborate on projects, track changes, and maintain code history. These systems help improve team coordination and ensure that projectfilesaresecurelystoredandeasilyaccessible[5],[6]. However,mostrepositoriesprimarilyfocusonstorageand collaborationfeaturesratherthanintelligentassistancefor debuggingandlearning.

1.2 Artificial Intelligence in Software Development

Artificial intelligence has significantly transformed many aspectsofsoftware engineering. Machinelearning models andlargelanguagemodelscananalyzeprogrammingcode, identifypatterns,anddetectpotentialerrors.Studieshave shownthatAI-basedtoolscanautomaticallygeneratecode suggestions, detect bugs, and recommend improvements, therebyreducingtheeffortrequiredformanualdebugging [7], [8]. The integration of such intelligent systems into development platformscan enhancecodingefficiencyand supportdevelopersinsolvingprogrammingproblemsmore effectively.

Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072

1.3 Need for an Intelligent Development Platform

Despite the availability of code repositories and AI tools, many developers still use separate systems for project managementandcodedebugging. Thisseparationcreates inefficienciesandincreasesthetimerequiredtoidentifyand correcterrors.Anintegratedplatformthatcombinesproject repository features with AI-powered code analysis can significantlyimprovedeveloperproductivityandlearning. Byproviding bothcollaborationand intelligentdebugging support in a single environment, developers can manage their projects more efficiently and resolve coding issues quickly[9],[10].

1.4 Overview of the Proposed System

The proposed SmartCodeHub platform aims to provide a secureandintelligentenvironmentforprojectcollaboration and code correction. The system allows users to register, uploadtheirprojects, andaccessprojectsshared byother users while maintaining strict access control. An administratormoduleensuressystemsecuritybyapproving userregistrationsandmanagingprojectcontent.Inaddition, theplatformintegratesanAICodeAssistantpoweredbythe Gemini model through the LangChain framework, which analyzesuser-submittedcodeandprovidescorrectedcode along with explanations. This integration transforms the platform into a smart development environment that supportsbothcollaborationandautomateddebugging.

2. PROPOSED SYSTEM

The proposed system, SmartCodeHub, is a web-based platform designed to integrate project repository management with an AI-powered code assistance mechanism. The system provides a secure environment where users can upload, manage, and share their projects whilealsoreceivingintelligentsupportfordebuggingand correctingcode.TheplatformisbuiltusingtheDjangoweb framework and integrates advanced artificial intelligence servicesthroughtheGeminiAPI.Thesystemensuressecure user authentication, role-based access control, and centralized administration for monitoring and managing useractivities.

The architecture of SmartCodeHub consists of several interconnectedmodulesincludingtheusermodule,admin module, project module, and AI chatbot module. These modulesinteractwithacentralizeddatabasetostoreuser information and project files. Additionally, the AI chatbot communicates with the Gemini API hosted in the cloud to analyzeandcorrectprogrammingcodesubmittedbyusers.

2.1 User Management Module

The User Management Module handles user registration, authentication,andaccountmanagement.Newusersmust firstregisterbyprovidingtheirbasicdetailssuchasname,

email,andpassword.Onceregistered,theaccountremains inactive until it is approved by the administrator. After approval, users can log in to the system and access their personalizeddashboard.

Fromthedashboard,userscanuploadprojects,viewtheir ownprojects,browseprojectsuploadedbyotherusers,and downloadsharedfiles.However,usersareallowedtoeditor delete only their own projects, ensuring proper access controlanddataownership.

2.2 Admin Management Module

The Admin Management Module provides centralized control over the system. The administrator logs into the platform using secure credentials and can monitor all registeredusersanduploadedprojects.Oneoftheprimary responsibilitiesoftheadministratorisapprovingnewuser registrations to maintain system security and prevent unauthorizedaccess.

Theadministratorcanalsoviewprojectsuploadedbyany user and delete inappropriate or unnecessary files if required.Thismoduleensuresthattheplatformmaintainsa structuredandcontrolledenvironmentforprojectsharing.

2.3 Project Repository Module

The Project Repository Module manages the storage and retrievalofprojectfileswithinthesystem.Userscanupload their project files, which are stored securely in the server and linked with their respective accounts in the database. The module allows users to view and download projects uploaded by others, encouraging collaboration and knowledgesharing.

Bymaintainingacentralizedrepository,thesystemensures thatprojectfilesareorganizedandeasily accessiblewhile maintainingproperaccessrestrictions.

2.4 AI Code Assistant Module

The AI Code Assistant Module is a key component of the SmartCodeHubplatform.Thismoduleenablesuserstopaste programmingcodecontainingerrorsorinefficienciesinto thesystem.ThesubmittedcodeisthensenttotheGeminiAI modelthroughtheLangChainframeworkforanalysis.

The AI model processes the code, identifies errors, and generatescorrectedcodealongwithexplanations.Thishelps usersunderstandtheirmistakesandimprovestheircoding skillswhilesignificantlyreducingdebuggingtime.

2.5 System Architecture

The system architecture of SmartCodeHub illustrates the interactionbetweentheuserinterface,applicationserver, database,andAIservices.Usersaccesstheplatformthrough awebbrowser,whichcommunicateswiththeDjango-based

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072

web interface. The server processes requests through various modules such as the user module, admin module, projectmodule,andAIchatbotmodule.

All user data and project information are stored in the centraldatabase.WhenauserinteractswiththeAIassistant, therequestissenttotheGeminiAPIthroughtheAIchatbot module,whichreturnsthecorrectedcodeandexplanation. Thisarchitectureensuresefficientcommunicationbetween systemcomponentsandenablesseamlessintegrationofAI capabilitiesintotheplatform.

Fig - 1. System Architecture of SmartCodeHub

The architecture demonstrates how different modules collaborate to provide secure project management and intelligent code correction within a unified development platform.

3. IMPLEMENTATION DETAILS

TheSmartCodeHubsystemisimplementedasaweb-based applicationusingtheDjangoframework,whichfollowsthe Model–View–Template (MVT) architecture. The platform integrates multiple modules including user management, projectrepositorymanagement,administrativecontrol,and anAI-poweredcodeassistant.Theimplementationensures secureauthentication,structureddatastorage,andseamless interactionbetweentheuserinterfaceandbackendservices. Thesystemalsointegratesartificialintelligencecapabilities through the Gemini 2.5 Flash model using the LangChain frameworkforintelligentcodeanalysisandcorrection.

3.1 Development Environment

The development of SmartCodeHub is carried out using modern web technologies to ensure reliability and scalability. The backend of the system is developed using Python and the Django web framework, which provides built-infeaturesforauthentication,databasehandling,and security. The frontend is developed using HTML, CSS, and basicJavaScript,providinganinteractiveanduser-friendly interface for users and administrators. The system uses SQLite as the database for storing user credentials and projectinformation.

The application is deployed locally using the Django developmentserver,andprojectfilesuploadedbyusersare stored in the server’s media directory. The integration of externalAIservicesisachievedthroughAPIcommunication withtheGeminimodel.

3.2 User Authentication and Registration

Theuserauthenticationmoduleisimplementedtomanage registration,login,andlogoutprocesses.Usersmustregister byprovidingtheirpersonaldetailssuchasname,email,and password.Afterregistration,theaccountremainsinactive untilitisapprovedbytheadministrator.

Django session management is used to maintain login sessionsandensuresecureaccesstosystemfeatures.Once approved by the administrator, users can log into the platform and access their personalized dashboard where theycanuploadandmanageprojects.

3.3 Project Repository Implementation

Theprojectrepositorymoduleallowsuserstoupload,view, anddownloadprojectfiles.Theuploadedfilesarestoredin theserver’smediafolderandarelinkedtotheuser’saccount in the database. Each project entry contains information suchasprojecttitle,filepath,uploaddate,anduserdetails.

UserscanviewtheirownuploadedprojectsthroughtheMy Projectssection,whiletheAllProjectssectionallowsthemto browse projects uploaded by other users. Access control mechanisms ensure that users can modify or delete only theirownprojects.

3.4 Admin Control Implementation

Theadministratormoduleisimplementedtomonitorand managetheentiresystem.Theadministratorlogsintothe system using predefined credentials and can view all registered users. One of the primary tasks of the administrator is approving newly registered users before theygainaccesstotheplatform.

Theadministratorcanalsoviewthelistofprojectsuploaded by any user and delete projects when necessary. This ensures that the system remains secure and free from unauthorizedorinappropriatecontent.

3.5 AI Code Assistant Integration

AsignificantfeatureofSmartCodeHubistheintegrationof theAICodeAssistant.Thismoduleisimplementedusingthe LangChainframework,whichactsasaninterfacebetween theDjangoapplicationandtheGemini2.5Flashmodel.

Users can paste programming code into the chatbot interface,andthesystemsendsthiscodetotheGeminiAPI foranalysis.TheAImodelprocessestheinputcode,detects possible errors, and generates corrected code along with

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072

explanations. The response from the AI service is then displayedontheuserinterface,allowinguserstounderstand theissueandapplythesuggestedimprovements.

3.6 File Storage and Database Management

ThesystemusesDjango’sfilehandlingmechanismstostore uploadedprojectfilessecurelyinthemediadirectory.The databasestoresimportantinformationsuchasuserdetails, projectmetadata,andapprovalstatus.

Bycombiningstructureddatabasestoragewithsecurefile management, the platform ensures efficient retrieval of projectdataandreliablesystemperformance.

4. RESULTS AND PERFORMANCE ANALYSIS

TheSmartCodeHubplatformwassuccessfullyimplemented and tested to evaluate its functionality, usability, and performance. The system integrates project repository managementwithanAI-poweredcodeassistant,providing users with a secure and intelligent development environment. The results demonstrate that the system effectively manages user authentication, project uploads, administrative control, and AI-based code correction. The performance evaluation focuses on system usability, response time, and accuracy of the AI-assisted debugging feature.

4.1 Home Page Interface

The home page serves as the entry point to the SmartCodeHubplatform.Itprovidesnavigationoptionsfor usersandadministratorstoaccessdifferentfunctionalities suchasuserregistration,userlogin,andadministratorlogin. Theinterfaceisdesignedtobesimpleandintuitive,enabling userstoquicklyunderstandthepurposeoftheplatform.The layout includes clearly labeled sections and buttons that guideuserstodifferentsystemmodules.

Thehomepagehighlightsthecorepurposeoftheplatform, which is to provide an intelligent repository for project collaborationandAI-poweredcodeassistance.

4.2 User Dashboard

Aftersuccessfulauthenticationandadministratorapproval, users can access their personalized dashboard. The dashboard provides options to upload new projects and explore various sections such as My Projects, All Users Projects, and the AI Code Assistant. The interface allows userstoeasilymanagetheirprojectsandaccesstheAI-based debuggingsystem.

The dashboard also includes a project upload form that allows users to add project files along with project titles. Uploadedprojectsarestoredsecurelyandcanbeaccessed laterforviewingordownloading.

4.3 System Performance

The performance of the system was evaluated based on response time, system reliability, and user interaction efficiency. The Django-based backend ensures efficient request handling and secure session management. File uploads and downloads are processed quickly due to optimized database queries and server-side storage mechanisms.

TheintegrationoftheGeminiAImodelthroughLangChain enables the AI Code Assistant to analyze and correct programmingcodeinrealtime.TheAImodulesuccessfully identifiescodingerrorsandgeneratescorrectedcodealong with explanations, improving the debugging process for users. Overall, the SmartCodeHub system demonstrates reliable performance and provides a user-friendly environmentformanagingprojectsandreceivingintelligent codingassistance.

5. CONCLUSIONS

The SmartCodeHub platform was developed to provide a unifiedenvironmentforprojectcollaborationandAI-based codeassistance.Thesystemsuccessfullyintegratesproject repository management with an intelligent debugging mechanism, allowing users to upload, manage, and share theirsoftwareprojectsina secureandorganizedmanner. Through the implementation of user authentication and administratorapprovalmechanisms,theplatformensures thatonlyauthorizeduserscanaccessandinteractwiththe

Fig - 2. SmartCodeHub Home Page

Fig - 3. User Dashboard

Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072

system,therebymaintainingdataintegrityandsecurity.One ofthekeycontributionsofthisprojectistheintegrationof anAICodeAssistantpoweredbytheGemini2.5Flashmodel usingtheLangChainframework.Thisfeatureenablesusers tosubmitprogrammingcodecontainingerrorsandreceive correctedcodealongwithclearexplanations.TheAI-driven approach significantly reduces the time required for debugging and helps users understand programming conceptsmoreeffectively.

The system also provides a structured administrative module that allows administrators to monitor user registrations, approve accounts, and manage uploaded projects.Thiscentralizedcontrolensuresthattheplatform operates efficiently and maintains high-quality project content.Additionally,theprojectrepositorymoduleallows users to explore and download projects shared by other users, promoting collaboration and knowledge sharing among developers. Overall, SmartCodeHub transforms a traditional project repository into an intelligent development platform by combining collaboration, secure access control, and AI-powered code correction. The successful implementation and testing of the system demonstrateitseffectivenessinimprovingcodingefficiency, simplifying debugging processes, and supporting collaborativesoftwaredevelopment.

6. FUTURE WORK

AlthoughtheSmartCodeHubplatformsuccessfullyintegrates project repository management with AI-powered code assistance,severalenhancementscanbeimplementedinthe future to improve its functionality and scalability. One possibleimprovementistheintegrationofversioncontrol features, which would allow users to maintain different versionsoftheirprojectsandtrackchangesovertime.This feature would make the platform more similar to professional development environments used in the softwareindustry.

Anotherimportantenhancementistheimplementation of real-timecollaboration,enablingmultipleuserstoworkon thesameprojectsimultaneously.Thiswouldsupportteambased development and improve communication among developers. Additionally, the AI Code Assistant can be extended to support multiple programming languages, provide code optimization suggestions, and generate automatic documentation or test cases for the uploaded code.Futureworkcanalsoincludeaddingadvancedsecurity mechanisms such as plagiarism detection, vulnerability scanning,andimprovedauthenticationmethodstoensure thesafetyofprojectfilesanduserdata.Thesystemcanalso beexpandedwithuserprofiles,projectratings,andfeedback systemstoenhanceinteractionamongusers.

Furthermore,deployingtheplatformoncloudinfrastructure anddevelopingamobile-friendlyinterfacewouldimprove accessibility and scalability. These improvements would

transformSmartCodeHubintoamorecomprehensiveand industry-readydevelopmentplatform.

REFERENCES

[1]I.Sommerville, Software Engineering,10thed.Boston, MA,USA:PearsonEducation,2016.

[2]R.S.PressmanandB.R.Maxim,SoftwareEngineering:A Practitioner’s Approach, 8th ed. New York, NY, USA: McGraw-HillEducation,2015.

[3]S.RussellandP.Norvig,ArtificialIntelligence:AModern Approach,4thed.PearsonEducation,2020.

[4] A. Vaswani et al., “Attention Is All You Need,” in Proc. Advances in Neural Information Processing Systems (NeurIPS),2017,pp.5998–6008.

[5] T. Brown et al., “Language Models are Few-Shot Learners,”inProc.NeurIPS,2020.

[6] A. Allamanis, M. Brockschmidt, and M. Khademi, “LearningtoRepresentProgramswithGraphs,”inProc.Int. Conf.onLearningRepresentations(ICLR),2018.

[7] S. Wang et al., “Deep Learning for Code Analysis: A Survey,”IEEETransactionsonSoftwareEngineering,vol.45, no.4,pp.345–365,2019.

[8] J. Jiang, Y. Liu, and Q. Xu, “Automatic Bug Fixing Using Neural Machine Translation,” in Proc. International ConferenceonSoftwareEngineering(ICSE),2018.

[9]GoogleDeepMind,“Gemini:AFamilyofHighlyCapable MultimodalModels,”GoogleAIResearchReport,2024.

[10] LangChain Developers, “LangChain: Framework for Developing Applications Powered by Language Models,” 2024.

[11]DjangoSoftwareFoundation,“DjangoWebFramework Documentation,”2024.

[12] GitHub Inc., “GitHub Documentation and Repository ManagementSystem,”2024.

[13] M. Fowler, Patterns of Enterprise Application Architecture.Boston,MA,USA:Addison-Wesley,2002.

[14]E.Gamma,R.Helm,R.Johnson,andJ.Vlissides,Design Patterns: Elements of Reusable Object-Oriented Software Boston,MA,USA:Addison-Wesley,1994.

[15] K. Murphy, Machine Learning: A Probabilistic Perspective.Cambridge,MA,USA:MITPress,2012.

[16]J.Han,M.Kamber,andJ.Pei,DataMining:Conceptsand Techniques,3rded.MorganKaufmann,2011.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 13 Issue: 03 | Mar 2026 www.irjet.net p-ISSN: 2395-0072

[17] H. Chen, C. Li, and Z. Huang, “AI-Based Code Error DetectionUsingDeepLearningTechniques,” International JournalofComputerApplications,vol.182,no.30,pp.15–21, 2022.

[18] S. Thrun and L. Pratt, Learning to Learn. Boston, MA, USA:Springer,2012.

[19]B.Meyer,Object-OrientedSoftwareConstruction,2nd ed.PrenticeHall,1997.

[20]T.H.Cormen,C.E.Leiserson,R.L.Rivest,andC.Stein, IntroductiontoAlgorithms,3rded.MITPress,2009.