Portfolio
Skills&Tools
Front-end
- HTML/CSS
- JavaScript
- React / Next.js
- Tailwind CSS
Back-end
- Node.js
- Express
- MySQL / PostgreSQL
- Prisma
Other
- C/C++ Programming
- Google Colab
- VS Code
- Postman
Projects
MoneyMap|FinanceTrackerApp

- DevelopedasecureandscalablefinancetrackerusingNextJS15andTailwindCSSwithClerkauthenticationandSupabasedatabase,deployedonVercel.
- IntegratedGoogleGeminiAPItoautomatetransactionreceiptparsingandpopulatedataaccurately.
- Implementedrobustbotprotection,ratelimiting,andemailvalidationviaArcjettoenhanceplatformsecurityandreliability.
- AutomatedtransactionalandmonthlybudgetreportswithResendforreal-timeusernotificationsandinsights.
- EnabledtransactiondataexportinbothCSVandPDFformatsforseamlessreportingandofflineaccess.
DashFlow|ProjectManagementApp

- DevelopedaJira-styleprojectmanagementplatformwithNextJS15andTailwindCSS,featuringanintuitiveKanbanBoardforreal-timetasktracking.
- Integratedadvancedanalyticsincludingmonthlyspendingtrends,networthtracking,budgetmanagement,cashflowanalysis,andfinancialhealthinsightstoempowersmarterdecision-making.
- Implementedsecureauthenticationandmulti-organizationrolebasedaccessusingClerk,enablingcollaborationacrossmultipleteamsandprojects.
- DesignedascalablebackendwithNeonDBtosupportdynamicissuemanagement,sprintplanning,andworkflowcustomization.
- DeployedtheapplicationseamlesslyviaVercel,ensuringfast,scalable,andreliableperformanceinproduction.
Education
| Course | Year | University / School | CGPA / % |
|---|---|---|---|
| B.Tech (MAC) | 2022 - 2026 | Netaji Subhas University of Technology | 7.88 |
| CBSE (Class XII) | 2022 | Sarvodaya Bal Vidyalaya No.2 | 92.8% |
| CBSE (Class X) | 2020 | Sant Nirankari Public School | 95.4% |
Extras
SentimentAnalysisinApacheSparkMLlib

- BuiltascalableMLpipelineonPySparkusingApacheSpark3.3.2andSparkMLlibfordistributeddataprocessingandclassificationtasks.
- Executeddataexploration,cleaning,andlabelencodingonaTwittersentimentdataset,filteringirrelevanttweetsandpreparingdataformodeling.
- Appliedtokenization,stop-wordremoval,andTF-IDFfeatureextractiontoconverttweettextintomeaningfulnumericalrepresentations.
- TrainedandevaluatedaLogisticRegressionmodelusingSparkMLlib,achievingperformanceinsightsviaaccuracy,precision,recall,andF1-scoremetrics.
LSTMTemperatureForecastModel
- CollectedandpreprocessedhourlyweatherdataforDelhi(2010–2025)byhandlinganomalies,scalingfeatures,andgeneratingtime-seriessequences.
- DevelopedadeeplearningmodelusingstackedLSTMlayerswithdropoutanddenselayers,builtwithTensorFlow/Kerastolearntemporalpatterns.
- Trainedthemodelover35epochsusingtheAdamoptimizerandMSEloss,validateditsperformanceusingRMSE,andvisualizedpredictionswithMatplotlib.
- Savedthetrainedmodelasa.h5fileforreuseandoutlinedfutureimprovementsincludinghyperparametertuningandintegrationofadditionalweatherfeatures.